Volume 60 Number 1

of the South African Institution of Civil Engineering March 2018

■■  easurement of free-flow conditions on multilane intercity

M
highways under heterogeneous traffic conditions
■■ I mpact of discounted professional fees on the
risk exposure of civil and structural engineering
services consultants in South Africa
■■ E valuation of locally available synthetic macro
fibres in a single-fibre pullout test in concrete
■■  he effect of aeration through an internal gallery of
T
a dam on the cavitation risk of Roberts splitters
■■  omparative analysis of methods of pile-bearing
C
capacity evaluation using CPT logs from tropical soils
■■  iscussion:
D
Planning for desalination in the context of the
Western Cape water supply system
Volume 60 No 1 March 2018 ISSN 1021-2019

Publisher
South African Institution of Civil Engineering
Block 19, Thornhill Office Park,
Bekker Street, Vorna Valley, Midrand, South Africa
Private Bag X200, Halfway House, 1685, South Africa
Tel +27 (0)11 805 5947/48, Fax +27 (0)11 805 5971
http://www.saice.org.za
civilinfo@saice.org.za

Contents
Editor-in-chief
Prof Gerhard Heymann
University of Pretoria
Tel +27 (0)12 420 3627
gerhard.heymann@up.ac.za
joint Editor-in-chief
Prof Chris Clayton 2 Measurement of free-flow conditions
University of Southampton
c.clayton@soton.ac.uk on multilane intercity highways under
MANAGING Editor heterogeneous traffic conditions
Verelene de Koker
Tel +27 (0)11 805 5947, Cell +27 (0)83 378 3996 A Boora, I Ghosh, S Chandra, K Rani DOI: 10.17159/2309-8775/2018/v60n1a1
verelene@saice.org.za
journal editorial Panel
Prof G Heymann – University of Pretoria 10 Impact of discounted professional fees on the
Prof CRI Clayton – University of Southampton
Prof Y Ballim – University of the Witwatersrand risk exposure of civil and structural engineering
Prof W Burdzik – University of Pretoria
Dr P Day – Jones & Wagener (Pty) Ltd services consultants in South Africa
Prof J du Plessis – University of Stellenbosch
Prof GC Fanourakis – University of Johannesburg P N Okonkwo, J Wium DOI: 10.17159/2309-8775/2018/v60n1a2
Prof M Gohnert – University of the Witwatersrand
Prof PJ Gräbe – University of Pretoria
Dr C Herold – Umfula Wempilo Consulting 21 Evaluation of locally available synthetic macro
Prof A Ilemobade – University of the Witwatersrand
Prof SW Jacobsz – University of Pretoria fibres in a single-fibre pullout test in concrete
Prof EP Kearsley – University of Pretoria
Prof C Roth – University of Pretoria C M Odendaal, A J Babafemi, R Combrinck, W I de Villiers, W P Boshoff
Prof W Steyn – University of Pretoria  DOI: 10.17159/2309-8775/2018/v60n1a3
Mr M van Dijk – University of Pretoria
Prof JE van Zyl – University of Cape Town
Prof C Venter – University of Pretoria
Prof A Visser – University of Pretoria
31 The effect of aeration through an internal gallery
Dr E Vorster – Aurecon South Africa (Pty) Ltd
Prof J Wium – University of Stellenbosch
of a dam on the cavitation risk of Roberts splitters
Prof A Zingoni – University of Cape Town G Calitz, G R Basson DOI: 10.17159/2309-8775/2018/v60n1a4
Prof M Zuidgeest – University of Cape Town
Peer reviewing
The Journal of the South African Institution of
Civil Engineering is a peer-reviewed journal
44 Comparative analysis of methods of pile-bearing
that is distributed internationally capacity evaluation using CPT logs from tropical soils
Design and reproduction
Marketing Support Services, Ashlea Gardens, Pretoria
I N Obeta, M E Onyia, D A Obiekwe DOI: 10.17159/2309-8775/2018/v60n1a5
Printing
Fishwicks, Pretoria
56 Discussion:
Papers for consideration should be submitted
online at: http://journal.saice.org.za
Planning for desalination in the context of the Western Cape
water supply system
The South African Institution of Civil Engineering accepts no
responsibility for any statement made or opinion expressed C L Blersch, J A du Plessis DOI: 10.17159/2309-8775/2018/v60n1a6
in this publication. Consequently, nobody connected with
the publication of this journal, in particular the proprietor,
the publisher and the editors, will be liable for any loss or
damage sustained by any reader as a result of his or her action 59 Thanks to our Referees!
upon any statement or opinion published in this journal.

© South African Institution of Civil Engineering 1

TECHNICAL PAPER Measurement of free-flow
Journal of the South African
Institution of Civil Engineering
conditions on multilane
ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 2–9, Paper 1678 intercity highways
AMARDEEP BOORA received a Bachelor of under heterogeneous
traffic conditions
Technology (BTech) degree from Kurukshetra
University, Kurukshetra, India, in 2010, and a
Master of Technology (MTech) from the
National Institute of Technology Kurukshetra
(NIT) in 2012 on the topic “Effect of fly ash on
the CBR value of different soils”. He is currently a PhD research scholar at
A Boora, I Ghosh, S Chandra, K Rani
the Indian Institute of Technology (IIT) Roorkee, India, and is working on
the topic “Operational performance measures for intercity roads”. He has
published more than ten research papers in reputed international journals
and conference proceedings. Vehicles travelling at their desired speed under free-flow conditions (FFC) on a highway without
Contact details getting influenced by other vehicle types travelling in the same lane exhibit a good level of
Department of Civil Engineering, Indian Institute of Technology Roorkee
service (LOS) on that particular highway facility. Few studies could be identified that provide
Uttarakhand, India 247667
T: +91 9416 803757, E: amardeepboora9@gmail.com guidelines on the measurement of FFC on multilane highways, including the latest version of
the United States’ Highway Capacity Manual (TRB 2010). As these guidelines are mostly based
DR INDRAJIT GHOSH is an Assistant Professor at
on homogeneous traffic conditions, they may or may not be applicable to heterogeneous
the Indian Institute of Technology (IIT) Roorkee,
India. He received his Bachelor of Engineering traffic conditions. Previously used methodologies (for two-lane and multilane highways)
(BE) and Master of Engineering (ME) degrees were examined in the current study, but were found to be unsuitable for multilane highways.
from the Bengal Engineering and Science Therefore a new measure, named speed difference (SD), was introduced in the present
University, Shibpur, India, in 2003 and 2005
study. SD is the difference in speed of two consecutive vehicles travelling in the same lane. A
respectively, and a Master of Science (MS) and PhD degrees in Civil Engineering
from Wayne State University, Detroit, Michigan, United States, in 2008 and relationship was established between SD and gap (the difference between two consecutive
2010. He is a Member of the Institute of Transportation Engineers (ITE) and an vehicles from back to front bumper while travelling in the same lane), instead of headway (the
Associate Member of the American Society of Civil Engineers (ASCE). He has
difference between two consecutive vehicles from back to back or front to front bumper while
teaching experience of more than six years, and has published several research
papers in reputed international journals and conference proceedings. He has travelling in the same lane). A gap value of 10 seconds was identified, beyond which vehicles
been involved in research on traffic safety and operations, traffic management, would be travelling in FFC. The guidelines proposed in the present study will be helpful for
and transportation data analysis using statistical and econometric methods. traffic engineers and planners in developing countries.
Contact details
Department of Civil Engineering, Indian Institute of Technology Roorkee
Uttarakhand, India 247667
T: +91 1332 285533, E: indrafce@iitr.ac.in INTRODUCTION interaction in the traffic stream. This
India has the second longest road network diversity in traffic characteristics makes
DR SATISH CHANDRA is currently the Director of
in the world – 5 263 001 km covering it more challenging to identify FFC as a
the Central Road Research Institute (CRRI), New
Delhi (since January 2016). Prior to that he was different road categories, namely national part of capacity and LOS measurements.
Professor of Transportation Engineering at the highways, state highways, other PWD This is particularly the case in develop-
Indian Institute of Technology Roorkee. roads, rural roads and urban roads, as per ing countries, e.g. India. If vehicles on a
Dr Chandra has published more than 200
the annual report of the Ministry of Road particular highway travel at speeds lower
research papers in journals and conference proceedings. He developed the
original concept of the Dynamic Passenger Car Unit for mixed traffic conditions. Transport and Highways (MoRT&H 2015). than their desired speed, it represents
He has guided 17 PhD and more than 100 MTech dissertations, and he was the As per the latest basic road statistics of a poor LOS (as the number of follow-
nodal officer for preparing the Indian Highway Capacity Manual. ers will be more). Followers are those
MoRT&H (2013), two-lane roads dominate
Contact details with 53% of the total road network, while vehicles that travel behind a slow-moving
Central Research Road Institute (CRRI)
Mathura Road, New Delhi, India 110025
multilane highways (four lanes or above) vehicle willingly or unwillingly (due to
T: +91 1126 848917 represent around 25%. the absence of suitable sight distance and
E: satisfce@gmail.com The latest statistics of MoRT&H (2015) gap to enable overtaking). Vehicles can
KAVITA RANI received her Bachelor of
show that vehicle ownership is continually travel at their desired speeds under FFC
Technology (BTech) from Kurukshetra increasing in India. This growth in the only (i.e. interaction between vehicles
University, Kurukshetra, India, in 2012, and a number of vehicles on the highways results will be minimum). FFC is the condition
Master of Technology (MTech) degree from
in increased congestion, making it difficult where vehicles travel at their desired
the Indian Institute of Technology Roorkee,
Roorkee, India, in 2017 on the topic for vehicles to travel at their desired speed speed without any influence from vehicles
“Walkability analysis of urban transport network”. (vehicles are forced to travel at a lower travelling in the same lane or adjacent
Contact details speed as they are influenced by the pres- lane. As it is clear that the FFC is directly
Department of Civil Engineering, Indian Institute of Technology Roorkee ence of vehicles in the adjacent lanes). At related to LOS, there is a need to identify
Uttarakhand, India 247667
the same time different vehicle categories the FFC for four-lane divided highways.
T: +91 9468 478687, E: kavita12aug@gmail.com
with different static and dynamic char- As there are no concrete definitions or
Keywords: free-flow condition, gap threshold, headway threshold, acteristics use the same roadway facility, guidelines to calculate free-flow speed
speed difference and this further increases the vehicular (FFS) on four-lane divided highways under

Boora A, Ghosh I, Chandra S, Rani K. Measurement of free-flow conditions on multilane intercity highways under heterogeneous
2 traffic conditions. J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1678, 8 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a1
heterogeneous traffic conditions, research-
ers globally are typically using the same
guidelines as suggested by the US HCM
(TRB 2010) (i.e. to calculate the FFS at low
traffic volumes of 1 400 PCU/‌h r or less).
But the FFS observed using this guideline
can be different from the actual FFS,
particularly at a different time interval
(e.g. from 11:00 to 13:00, or from 16:00
to 18:00). It was concluded earlier (Dixon
et al 1999) that the HCM guidelines
could not be used to calculate FFS. Tseng
et al (2005) calculated FFS by taking
those vehicles that were travelling with a
headway value greater than 5 seconds as
vehicles travelling in FFC without getting
influenced by other vehicles present on
the same roadway facility. Arasan and
Dhivya (2010) conducted a study to deter-
mine the concentration of heterogeneous
traffic conditions on multilane highways.
The FFS in their study was calculated at
low traffic volumes, as the movement of
a vehicle was not influenced by the pres-
ence of another vehicle in the same lane
or in the adjacent traffic lane on the same
highways facility. Alternatively, Semeida
(2013) calculated the FFS of vehicles that
were travelling with a headway value
greater than 8 seconds, while Robertson et
al (2014) described free-moving vehicles
as those vehicles that were travelling with
headway values of 5 seconds, and tailway
values of 3 seconds. In most of these stud-
ies, the FFS was calculated by adopting the
guidelines from what had previously been
proposed for two-lane highways. In the lit-
erature only a single study was found that
related to the car-following interaction.
Al-Kaisy and Karjala (2010) used eight
study sites in Montana State to examine
car-following behaviour, as well as to
calculate the FFS on two-lane roads and
four-lane roads. It was concluded from the
study that after 6 seconds headway thresh-
old, interaction between different vehicles
diminishes and vehicles will travel at their
FFS or in FFC on two-lane roads. In the
case of four-lane roads no pattern was
observed, and it was concluded that the
vehicle interaction on multilane highways
under low traffic volume is very difficult
to observe. An earlier study (Fitzpatrick et
al 2005) was conducted to calculate FFS
in suburban/urban and rural areas by tak-
ing the average speed of the vehicles that
were travelling with 5 seconds headway
and 3 seconds tailway. In another study
(Figueroa & Tarko 2005) a headway value
equal to or greater than 5 seconds was
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
identified, beyond which vehicles were
found to travel in FFC on the tangent
segment and horizontal curve of two-lane
rural highways. Earlier, Al-kaisy & Durbin
(2008) proposed a headway threshold
value ranging from 5 to 7 seconds, beyond
which interaction between vehicles dimin-
ished, i.e. the vehicles were travelling in
FFC. Another study (Lobo et al 2011) was
conducted on Portuguese two-lane rural
roads to calculate the FFS, and a free gap
of 6 seconds was found appropriate to
calculate the FFS, as vehicles were travel-
ling in a free condition beyond that gap
value. In another study, Hashim (2011)
analysed speed characteristics on two-lane
rural highways in Egypt. From the study
it was revealed that 85th percentile speed
did not vary after a critical headway value
of 5 seconds, which indicates that vehicles
will travel in FFC after 5 seconds headway
value, and FFS can be calculated for head-
way value equal to or more than 5 seconds
between two consecutive vehicles. In
contrast, Vogel (2002) characterised free
vehicles in urban areas by collecting the
data on a single location, which was a
four-leg two-way stop-controlled intersec-
tion with a 6 seconds headway, beyond
which interaction between vehicles was
diminished, while another study (Radwan
& Kalevela 1985) suggested a headway
greater than 9 seconds to identify FFC.
But these guidelines are based on homo-
geneous traffic conditions and may or may
not be applicable to heterogeneous traffic
conditions, as the use of these guidelines
can give misleading results. Measurement
of FFS is required to exhibit the LOS pro-
vided to the road users, i.e. vehicles travel-
ling at their desired speed in FFC exhibit
good LOS at that particular site, otherwise
there is a need to widen (i.e. in the case
of congestion) the roadway. Besides this,
once the FFS has been obtained for the
different study sites, a capacity model can
be developed by using the ratio of FFS to
the capacity values of different sections,
Table 1 Vehicle composition with traffic flow for each study site
Peak time Total
Site Chainage
period vehicle
Site 1 55 km 16:00 – 18:00 5 298
Site 2 70 km 10:00 – 12:00 3 360
Site 3 23 km 14:00 – 16:00 3 188
Site 4 5 km 10:00 – 12:00 2 634
Site 5 30 km 11:00 – 13:00 1 530
which makes the capacity estimation easy
in the field. Therefore, instead of apply-
ing the guidelines directly to four-lane
divided multilane highways, all the previ-
ously used methodologies for multilane
and two-lane highways are examined in
this study.
DATA COLLECTION
Different study sites were identified,
namely NH–3 (Mumbai–Nashik Highway)
for Site 1, three sections (Sites 1, 2 and 3)
from NH–58 and SH–30 (Ujjain–Indore
Highway), and Site 5 which was selected
from different regions of India. All of the
study sites are located on level terrain.
These study sites were located far away
from any influence factors such as curves,
traffic control devices and access points
up to 500 m in both directions. The data
collection was conducted during daylight
hours (between 08:00 and 18:00) on typical
weekdays using a videography technique. A
longitudinal trap of 50–60 m was made on
the road surface for measuring the speed.
A video camera was positioned in such a
manner that each vehicle could be identi-
fied easily along the entire trap, especially
at the entry and exit points. Data extrac-
tion was done manually by playing the
video in the laboratory on a large television
screen for a different time interval (peak
hours only) as per the study requirements,
as shown in Table 1. Consequently traffic
volume, composition, vehicular speed, 85th
percentile speed, and the travel mean speed
of both the traffic stream and every vehicle
category were calculated. The headway
and gap values between two consecutive
vehicles travelling in a lane were extracted
from the collected video data. Vehicles
were categorised into four categories,
namely passenger car (PC), motorised two-
wheeler (2W) (motor cycle), auto-rickshaw
(AR) (three-wheeler) and heavy vehicle
(HV). This was done for all five study sites,
as shown in Table 1.
Flow
PC (%) 2W (%) AR (%) HV (%)
(veh/hr)
42.13 34.47 8.06 15.34 2 649
45.21 36.31 4.85 13.63 1 680
46.74 34.35 6.43 12.48 1 594
44.67 35.56 6.11 13.66 1 317
34.23 53.82 2.35 9.60 765
3
 Site 4 Site 5
65 65

64 64
ATS (km/h)

ATS (km/h)
63 63

62 62

61 61
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(a) Headway threshold (seconds) (b) Headway threshold (seconds)

Site 4 Site 5
80 80
85th percentile speed (km/h)

85th percentile speed (km/h)

79 79
78 78
77 77
76 76
75 75
74 74
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(c) Headway threshold (seconds) (d) Headway threshold (seconds)

 elationships of ATS and 85th percentile speed with headway equal to or greater than a specific threshold value
Figure 1 R

Other vehicle categories, such as
animal-drawn vehicles and bicycles, were
not considered in the study. The traffic
volume was found to vary from 765 veh/hr
to over 2 649 veh/hr. The variation in traf-
fic composition is shown in Table 1, which
indicates a high proportion of PC on all
the study sites, except Site 5 where 2W was
dominating the traffic.
MEASUREMENT OF
FREE-FLOW CONDITION
In the present study, previously used meth-
odologies that were proposed for two-lane
highways in order to identify the FFC, are
tried for multilane highways under hetero­
geneous traffic condition. As discussed
earlier, Al-Kaisy and Karjala (2010) carried
out a study to examine the car-following
interaction on two-lane highways, as well
as on multilane highways. A headway
threshold value of 6 seconds was identified,
beyond which all vehicles were travelling in
FFC, while on multilane highways contra-
dicting results were obtained, i.e. average
travel speed (ATS) was decreasing with
headway threshold value.
In another study Hashim (2011)
identified a headway value of 5 seconds,
beyond which vehicles were identified as
free-moving vehicles (i.e. travelling with
equal to or greater than 5 seconds headway
value), while Al-kaisy and Durbin (2008)
identified FFC with headway threshold
values ranging from 5 to 7 seconds. In
the present study, same methodologies
were examined by establishing different
relationships for headway threshold with
ATS and 85th percentile speed of the
traffic stream (i.e. including all vehicle
categories) for all the study sites. Here the
ATS of vehicles travelling with the same or
greater than a specific headway threshold
value were plotted at all the study sites.
Figure 1 shows the same relationship for
Site 4 and Site 5. It is a general hypothesis
that ATS will increase with an increase in
headway value and become constant after
a critical headway threshold value which
exhibits FFC. At study Site 4, the ATS was
found to decrease with increasing headway
­thres­hold value, which contradicts the gen-
eral expectation, as shown in Figure 1(a),
while in the case of Site 5, ATS was found
to increase with headway threshold value,
as shown in Figure 1(b). But none of these
graphs showed any headway threshold
value beyond which vehicles could be
characterised as free-moving vehicles as
obtained in a previous study (Al-Kaisy and
Karjala 2010).
The main reason for the contradicting
results could be the difference in traffic
volumes, as shown in Table 1. Another
reason could be driver behaviour and age
(e.g. an aggressive driver). Following this,
other relationships were established at all
the study sites for 85th percentile speed
with a headway threshold value in order to
identify the FFC. The same hypothesis was
made for 85th percentile speed with a head-
way threshold value as made in the case of
the ATS. However, no trend was observed
from any graph, as shown in Figures 1(c)
and 1(d).
Similar relationships were developed for
all the study sites and similar results were
obtained as shown in Figure 1 for Site 4 and
Site 5. A total of about 15 369 headway and
gap values were observed during the analy-
sis. Consequently, a new definition for head-
way threshold value was introduced in the
present study in which ATS and 85th per-
centile speeds of vehicles travelling with less
than a specific headway value were plotted.
Figure 2 shows such a relationship for the
same study sites (Site 4 and Site 5). However,
after plotting the graphs, the ATS was found
to decrease with the headway threshold
value at Site 4, which contradicts the general
hypothesis (i.e. due to the variation in vehi-
cle length which also affects the headway
value and result), while at Site 5 the relation-
ship of the ATS with headway threshold
value did not show any trend, as the ATS
was found to increase with the headway
threshold value. On the other hand, 85th
percentile speed also did not show any trend
as per the general hypothesis. At Site 5,
85th percentile speed was found to increase
with the headway threshold value, while no

4 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 Site 4 Site 5
67 60

66 59
ATS (km/h)

ATS (km/h)
65 58

64 57

63 56
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(a) Headway threshold (seconds) (b) Headway threshold (seconds)

Site 4 Site 5
78 72
85th percentile speed (km/h)

85th percentile speed (km/h)

77 71
76 70
75 69
74 68
73 67
72 66
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(c) Headway threshold (seconds) (d) Headway threshold (seconds)

 elationships of the ATS and 85th percentile speed, with headway less than a specific threshold value
Figure 2 R

trend was observed at Site 4, as shown in Penmetsa et al (2015) proposed a gap value pattern shows the high frequency at low
Figures 2(d) and 2(c) respectively. It was con- of 2.6 seconds to differentiate follower and gap values, which decreased gradually
cluded from Figures 1 and 2 that headway not-following vehicles. Consequently, gap with an increase in the gap value between
cannot be used to identify the FFC in which value was used instead of headway to mea- vehicle arrivals at both study sites. From
vehicles would travel at their desired speed. sure the FFC in the present study. these distribution graphs, two regions were
Difference in the traffic characteristics of identified. The first region was identified
developing and developed countries (i.e. Characterisation of gap distribution where a sharp decrease in frequency was
heterogeneous versus homogeneous) can As gaps do not vary with vehicle length, it observed, which was evident in the first
be a reason why previous methodologies was used in the present study to identify five gap values, while the second region
seem inappropriate for heterogeneous traffic the FFC. Initially, in order to understand was identified after the first five gap values
conditions. the vehicle interaction on multilane high- where a steady decrease in frequency was
Earlier, Van As (2003) and Penmetsa ways, gap distributions were investigated at observed. The kink in the fictional curve
et al (2015) also observed that US HCM all the study sites. A frequency histogram is found between two regions some place
methodologies are not appropriate for showing the gap distribution at two of after a gap value of 10 seconds. Earlier,
heterogeneous traffic conditions. Van the study sites (Site 4 and Site 5) is shown Miller (1961) identified this boundary with
As (2003) proposed a headway value of in Figure 3. As the traffic volumes varied a headway value of 6 seconds at two-lane
3.5 ­seconds to identify the following from low to a high on all the study sites, highways. The same histograms were plot-
vehicles instead of 3.0 seconds as proposed data up to a 30 seconds gap value was used ted for the remaining study sites, which
by the HCM 2000 (TRB 2000), while to analyse the gap distribution. The general also exhibited the same trend.

Site 4 Site 5
500 484 150 145

391 120
400
331
100 88
Frequency

Frequency

300
250 75 72
211 69 70 65 66
61
200 50 53
50 45
130 39
91 81 33
100 70 63 53
32 24 25 16
0 0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
(a) Gap (seconds) (b) Gap (seconds)

Figure 3 F requency histogram of gap distribution on multilane highways

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 5
 Site 4 Site 5
65 65

64 64
ATS (km/h)

ATS (km/h)
63 63

62 62

61 61
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(a) Gap threshold (seconds) (b) Gap threshold (seconds)

Site 4 Site 5
79 79
85th percentile speed (km/h)

85th percentile speed (km/h)

78 78

77 77

76 76

75 75

74 74
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(c) Gap threshold (seconds) (d) Gap threshold (seconds)

 elationships of ATS and 85th percentile speed with gap equal to or greater than a specific threshold value
Figure 4 R

Thereafter different graphs were plotted
for ATS and 85th percentile speeds with
a gap equal to or greater than a specific
gap threshold value as plotted earlier,
with a headway threshold to identify the
FFC as shown in Figure 4. Similar results
were obtained with the gap threshold as
in Figure 1. Later on, these graphs were
plotted again with the gap less than a
specific gap threshold value, as shown in
Figure 5. As is evident from Figure 5(a) in
the case of Site 4, the ATS did not show
any trend, while at Site 5 the ATS was
found to increase with gap threshold value,
as shown in Figure 5(b). On the other hand,
85th percentile speed also did not show
any trend with the gap threshold value.
After analysing different relationships
of ATS and 85th percentile speed with
headway and gap threshold values, it was
concluded that previously used parameters
to identify the FFC on multilane highways
under heterogeneous traffic condition
cannot be used.

Site 4 Site 5
79 60

78 59
ATS (km/h)

ATS (km/h)

77 58

76 57

75 56
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(a) Gap threshold (seconds) (b) Gap threshold (seconds)

Site 4 Site 5
78 74
85th percentile speed (km/h)

85th percentile speed (km/h)

72
77 70
68
76
66

75 64
62
74 60
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(c) Gap threshold (seconds) (d) Gap threshold (seconds)

 elationships of ATS and 85th percentile speed with gap less than a specific threshold value
Figure 5 R

6 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 Site 4 Site 5
13 13
12 12
11 11
SD (km/h)

SD (km/h)
10 10
9 9
8 8
7 7
6 6
5 5
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
(a) Gap threshold (seconds) (b) Gap threshold (seconds)

Figure 6 R
 elationships between SD with gap less than a specific threshold value

Because of that, a new measure, named
speed difference (SD), was introduced in
the present study. SD is the difference in
speed of two consecutive vehicles in the
same direction of travel in the same lane.
Same relationships were established again
for SD with gaps less than a specific gap
threshold value at both study sites, as
shown in Figure 6.
After analysing at the microscopic level
it was observed that the SD between two
consecutive vehicles increased continuous-
ly up to a critical gap value, beyond which
this SD curve becomes almost flat. This
gap value was observed to be 10 seconds,
which indicates that a vehicle travelling
with a gap value greater than 10 seconds
can be considered as a free-moving vehicle
travelling at its desired speed under FFC.
During the study, two types of drivers
were identified – those who travel in fol-
lowing mode by choice and who can be
termed as “happy to follow”, and those who
are forced to travel in following mode due
to overtaking restrictions. A gap value of
10 seconds was identified, beyond which
vehicles would travel as free-moving
vehicles at their desired speeds on four-lane
divided highways under heterogeneous
traffic condition.
FOLLOWING BY CHOICE
PHENOMENA
There is a possibility of free-moving
vehicles below the 10 seconds gap value
(i.e. happy to follow). So, there is a need to
identify a critical gap value below which all
vehicles would travel in following mode.
A histogram of a whole data sample was
plotted, as shown in Figure 7(a) for study
Site 4. Simultaneously, a normal curve of
free-flowing vehicles (i.e. travelling with
equal to or greater than 10 seconds gap
value) was superimposed on the histogram
to identify an SD limit beyond which a
vehicle would be in FFC, as shown in
Figure 7(a). A superimposed normal curve
intersects the histogram and exhibits an
SD limit of –2.55 to +9.64 km/h, with a
maximum number of vehicles travelling in
the field. After observing the field data, an
SD limit range of –7 to +15 km/h seemed
more appropriate to identify the followers
and non-followers. Later, an acceptance
curve method was used to identify a
critical gap value beyond or below which
vehicles would travel as free vehicles and
following vehicles respectively, as shown
in Figure 7(b). A critical gap value of
4.2 seconds was identified, beyond which
the probability of vehicles travelling in FFC
would increase, as shown in Figure 7(b).
Similarly, different SD limit ranges and
critical gap values were identified on
all the remaining study sites, as shown
in Table 2. Due to the fact that a driver
expects high speed at four-lane divided
highways (as there is no interaction from
the opposite direction), high ranges of SD
were observed in the field, as shown in the
second column of Table 2 on page 8.
DEVELOPMENT OF MODEL
It is difficult for traffic engineers and plan-
ners to repeat the full set of procedures in
the field in order to identify followers and
non-followers. Consequently, a model was
developed by establishing the relationship
between the traffic volume and the gap

700 1.0
0.9
600
Normal curve of
Probability of not following

0.8
free‑moving vehicles
500 0.7
Frequency

400 0.6
0.5
300 0.4
200 0.3
0.2
100
0.1
0 0
–68 –58 –48 –38 –28 –18 –8 2 12 22 32 42 52 62 0 1 2 3 4 5 6 7 8 9 10
(a) SD (km/h) (b) Gap (seconds)

Figure 7 ( a) Histogram of the speed difference between consecutive vehicles; superimposed is the normal curve of relative speed of non-following
vehicles; (b) Probability of not following at Site 4

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 7
Table 2 Different SD limit and critical gap values observed in the field values observed on all the study sites, as
SD limit Traffic volume Critical gap value shown in Figure 8. An exponential trend
Study sites
(km/h) (veh/h) (seconds) was found to be the best fit and can be
expressed in mathematical form as shown
Site 1 –4.16 to +15 2 649 2.2
in Equation 1.
Site 2 –6.8 to +11.01 1 680 3.37
Y = 7.503854e–0.0004688x(1)
Site 3 –3.31 to +6 1 594 3.4

Site 4 –2.55 to +9.64 1 317 4.2 Where:

Y = Gap value
Site 5 –2.19 to 11.32 765 5.28
x = Traffic volume observed in the field

Validation of model
6 Thereafter a new study site located on the
Dewas to Indore State Highway (SH)–18
5 was selected to validate the model. The
same procedure was used to identify the
4 critical gap value for SH–18 in the field that
Gap (seconds)

was used for Site 4, as shown in Figure 9.

3
Table 3 shows the different gap values
y = 7.50385447e–0.00046881x observed in the field, and from Equation 1.
2 R2 = 0.9910018 After comparing the gap value obtained
from the field and from the model, an error
1
of 3.02% was observed, which appears
acceptable. By using this exponential
0
500 1 000 1 500 2 000 2 500 3 000 model, users or traffic planners can iden-
Traffic volume (vehicles/hour) tify the followers and non-followers easily,
but this model is valid only for the particu-
Figure 8 V
 ariation in gap values in traffic volumes at all the study sites lar range of traffic volumes as shown in
Table 3.

1.0
SUMMARY OF FINDINGS
0.9
AND RECOMMENDATIONS
In this study FFC was identified on four-
0.8
lane divided multilane highways under
0.7 heterogeneous traffic conditions. The most
Probability of not following

important findings of the study are the

0.6 following:
■■ It was concluded that previously used
0.5
parameters (i.e. ATS and 85th percentile
speed) and methodologies for two-lane
0.4
highways cannot be used for multi-
0.3 lane highways under heterogeneous
traffic conditions.
0.2 ■■ It was also concluded that the headway
cannot be used for multilane highways
0.1
under heterogeneous traffic conditions
to identify FFC, as it is dependent on
0
0 1 2 3 4 5 6 7 8 9 10 vehicular length.
Gap (seconds) ■■ Gap is found appropriate instead of
headway for heterogeneous traffic
Figure 9 V
 ariation in gap values in traffic volumes at all the study sites conditions, as it does not vary due to
vehicle length.
Table 3 Difference in gap values observed in the field and from the model ■■ A new measure named speed difference

Gap value (second) (SD) was introduced in this study and

Traffic
% found to be appropriate in identifying
Study site volume Observed in the field Observed from Error
(km/h) as shown in Figure 9 Equation 1 the FFC.
■■ An SD limit ranged from –7 to +15 km/h
Site 6 (Dewas to Indore) 1 120 4.3 4.43 3.02
and a gap value of 10 seconds were

8 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 identified. This gap value was used
to identify the followers on four-lane
divided highways under heterogeneous
traffic conditions.
■■ Consequently, an exponential model
was also developed to identify the criti-
cal gap value in the field.
■■ The empirical observation revealed that
the vehicle interaction diminished after
a 10 seconds gap value on four-lane
divided multilane highways.
■■ A significant proportion of drivers give
preference to maintaining a shorter gap
while travelling on four-lane divided
highways, irrespective of passing (over-
taking) restrictions.
This analysis of four-lane divided highways
under heterogeneous traffic conditions
can be further extended by using compre-
hensive field data from different parts of
the country. Guidelines proposed in the
present study to identify the FFC on four-
lane divided highways would help traffic
engineers and planners to calculate the FFS
under heterogeneous traffic conditions.
Furthermore, different LOS thresholds
can be calibrated for four-lane divided
highways under heterogeneous traffic
conditions using the guidelines proposed in
this study.
ACKNOWLEDGEMENTS
The work reported in this paper is part of
an ongoing research project (Development
of Indian Highway Capacity Manual
(INDO-HCM)) sponsored by CSIR-CRRI,
New Delhi, India. The financial assistance
provided by the sponsoring agency for traf-
fic studies is gratefully acknowledged.
REFERENCES
Al-Kaisy, A & Durbin, C 2008. Evaluating new
methodologies for estimating performance on
two-lane highways. Canadian Journal of Civil
Engineering, 35(8): 777–785.
Al-Kaisy, A & Karjala, S 2010. Car-following
interaction and the definition of free-moving
vehicles on two-lane rural highways. Journal of
Transportation Engineering, 136(10): 925–931.
Arasan, V T & Dhivya, G 2010. Methodology for
determination of concentration of hetrogeneous
traffic. Journal of Transportation Systems
Engineering and Information Technology, 10(4):
50–61. Available at: http://dx.doi.org/10.1016/
S1570-6672(09)60052-0.
Dixon, K, Wu, C H, Sarasua, W & Daniel, J 1999.
Estimating free-flow speeds for rural multilane
highways. Transportation Research Record,
1678: 73–82.
Figueroa, A M & Tarko, A P 2005. Speed factors on
two-lane rural highways in free-flow conditions.
Paper presented at the 84th Annual Meeting of
the Transportation Research Board, 9–13 January,
Washington, DC.
Fitzpatrick, K, Miaou, S P, Brewer, M, Carlson, P
& Wooldridge, M D 2005. Exploration of the
relationships between operating speed and
roadway features on tangent sections. Journal of
Transportation Engineering, 131(4): 261–269.
Hashim, I H 2011. Analysis of speed characteristics for
rural two-lane roads: A field study from Minoufiya
Governorate, Egypt. Ain Shams Engineering
Journal, 2(1): 43–52. Available at: http://dx.doi.
org/10.1016/j.asej.2011.05.005.
Lobo, A, Jacques, M, Rodrigues, C & Couto, A 2011.
Free-gap evaluation for two-lane rural highways.
Transportation Research Record, 2223: 9–17.
Miller, A 1961. A queueing model for road traffic flow.
Journal of the Royal Statistical Society. Series B
(Methodological), 23(1): 64–90.
MoRT&H (Ministry of Transport & Highways) 2013.
Basic Road Statistics of India 2011–12, p 109.
New Delhi, India: MoRT&H.
MoRT&H (Ministry of Transport & Highways)
2015. Annual Report 2014–15. New Delhi, India:
MoRT&H.
Penmetsa, P, Ghosh, I & Chandra, S 2015. Evaluation
of performance measures for two-lane intercity
highways under mixed traffic conditions. Journal of
Transportation Engineering, 141(10): 04015021.
Radwan, A E & Kalevela, S 1985. Investigation of
the effect of change in vehicular characteristics
on highway capacity and level of service.
Transportation Research Record, 1005: 65–71.
Robertson, J, Fitzpatrick, K, Park, E & Iragavarapu,
V 2014. Determining level of service on freeways
and multilane highways with higher speeds.
Transportation Research Record, 2461: 85–93.
Semeida, A M 2013. New models to evaluate the level
of service and capacity for rural multilane highways
in Egypt. Alexandria Engineering Journal, 52(3):
455–466. Available at: http://dx.doi.org/10.1016/j.
aej.2013.04.003.
TRB (Transportation Research Board) 2000. HCM
(Highway Capacity Manual) 2000. Washington, DC:
National Research Council.
TRB (Transportation Research Board) 2010. HCM
(Highway Capacity Manual) 2010. Washington, DC:
National Research Council.
Tseng, P, Lin, F & Shieh, S L 2005. Estimation of
free-flow speeds for multilane rural and suburban
highways. Transportation, 6: 1484–1495.
Vogel, K 2002. What characterizes a “free vehicle”
in an urban area? Transportation Research Part F:
Traffic Psychology and Behaviour, 5(1): 313–327.
Van As, C 2003. The development of an analysis
method for the determination of level of service
of two-lane undivided highways in South Africa.
Project Summary. Pretoria: South African National
Roads Agency Limited (SANRAL).

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 9
TECHNICAL PAPER Impact of discounted
Journal of the South African
Institution of Civil Engineering
professional fees on the risk
ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 10–20, Paper 1549 exposure of civil and structural
PATRICK NWABUEZE OKONKWO is a PhD engineering services
consultants in South Africa
candidate at the department of Civil
Engineering (Construction Engineering and
Management Division), Stellenbosch
University, where he is conducting research
in the area of construction risk
management. He received his M Eng Civil
P N Okonkwo, J Wium
Engineering degree from Stellenbosch
University, and his B Eng (Civil) degree from the University of
Benin, Nigeria.

Contact details
Department Civil Engineering
Stellenbosch University
Private Bag X1
Matieland 7602
South Africa
E: patrickezeokonkwo@gmail.com
PROF JAN WIUM Pr Eng (MSAICE) is
professor in the Murray & Roberts Chair of
Construction Engineering and Management
in the Department of Civil Engineering at
Stellenbosch University. He obtained a BSc
(Eng) and MSc (Eng) from the University of
Pretoria, and a PhD from the Swiss Federal
Institute of Technology in Lausanne. He
worked as a consultant for 20 years before joining Stellenbosch University
in 2003. After first researching the behaviour of concrete structures and
seismic analysis of structures, he now focuses his research on the
management and initiation of multidisciplinary projects.
The abolition of mandatory fee scales and the prevalence of lowest-cost bidding for the
procurement of consulting services in South Africa have seen engineering services consultants
compete based on price for engineering contracts. Discounts benchmarked against ECSA
(Engineering Council of South Africa) professional fees guidelines demanded by clients have
resulted in declines in professional fees over the years. The capacity to deliver professional
services that are of such high quality that it meets the client’s expectation, professional and
ethical standards when working at low fees is one of the biggest challenges facing consulting
professionals today. This research studied the risks encountered by civil and structural
engineering services consultants and the impact of discounted professional fees on their risk
exposure. The study included a review of literature, discussions with practising engineers and
a questionnaire survey of 23 practising consulting engineers representing small, medium and
large consulting engineering firms. A key finding of this study is that discounted fees accentuate
several project level risks and create organisational level risks for the consulting engineering
professional. The implications of these are discussed, and recommendations for improving the
industry put forward.

Contact details BACKGROUND (CESA 2013). The implication of this is

Department Civil Engineering
Stellenbosch University
It is widely known that consulting engi- that contracts are awarded to the firm
Private Bag X1 neers are liable for the professional advice that quotes the lowest price to provide the
Matieland 7602 they give to their clients. However, within service (CESA 2014).
South Africa
the South African environment, the abil- Professional bodies such as Consulting
ity of consulting engineers to continually Engineers South Africa (CESA) have,
provide sound professional advice in the however, maintained that consulting engi-
face of declining professional fees in the neering services is not a commodity, and
industry has been a cause for concern as such the use of competitive tendering
amongst stakeholders and academics alike. procurement methods based on price is
Weidemann (2014) pointed out that engi- inappropriate (Thela 2014). This argument
neers in South Africa are struggling “in an is based on the fact that, while it is possible
industry racked by discounting, tendering, to draft specifications against which the
lack of knowledge and skills.” quality of commodities (physical goods)
The decline in professional fees in will be evaluated, such specifications
the country over the years can be partly cannot be drawn up easily for consulting
attributed to competitive tendering and the services (Davies 2006).
practice of discounting professional fees Lu et al (2013) provide a detailed
benchmarked against the professional fee review of previous research carried out
guidelines published by the Engineering in the field of construction professional
Council of South Africa (ECSA) (CESA services (CPS), and segmented the focus
2007). Given that quality and cost-based areas into the following subject categories:
selection (QCBS) methods of tender evalu- competitiveness of CPS firms, character-
ation are used for public procurement by istics of CPS firms, impact of information
the government (the largest client for these technology on CPS firms, management
Keywords: consulting engineer, professional fees, discounts, risk, professional services in South Africa), cost practices of CPS firms and procure-
risk management, South Africa plays a crucial role in contract allocation ment methods of the CPS industry.

Okonkwo PN, Wium J. Impact of discounted professional fees on the risk exposure of civil and structural engineering services consultants in South Africa.
10 J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1549, 11 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a2
This submission by Lu et al (2013) and a
perusal of literature reveal that the subject
of risk, as it relates to the CPS industry,
has received very little attention.
Also, while the inherent risks for
clients who engage professional ser-
vices providers at low fees have been well
researched and identified to include low
quality of service resulting in increased
life cycle cost of the project, and project
delays resulting from reworks and poor
contract documentation (Hoxley 2000;
Ling 2004; Love & Edwards 2004), perusal
of the literature revealed that the implica-
tion of low fees on the risk exposure of
the consulting engineering professional
is highly under-researched. This study
attempts to fill this research gap by pursu-
ing the following objectives:
■■ To obtain information on project
and organisational risks particular to
the civil and structural engineering
services consultants.
■■ To obtain information on the extent of
discounting practices amongst consult-
ing engineering firms in South Africa.
■■ To determine the impact of discounted
fees on project and organisational
risk factors by ranking them accord-
ing to how they are impacted by
discounted fees.
■■ To identify measures adopted by con-
sulting engineers to manage and miti-
gate the influence of low professional
fees on their risk exposure.
LITERATURE REVIEW
Procurement and remuneration for
consulting engineering services
In South Africa, public sector clients
account for the greater percentage of
professional fees earned by consulting
engineering firms. In 2013, for example,
54% of all fees earned in the first half of
the year were from public sector clients
(CESA 2013). The procurement of consult-
ing engineering services in South Africa
by this category of clients is governed by
several procurement legislations, such
as the Preferential Procurement Policy
Framework Act (PPPFA), Public Finance
Management Act (PFMA), Municipal
Finance Management Act (MFMA), Broad-
Based Black Economic Empowerment
(BBBEE), various Treasury (government
department) and Construction Industry
Development Board (CIDB) procurement
guidelines (CESA 2014). These legislations
stipulate the use of quality and cost-
based selection (QCBS) methods for the
evaluation of tenders. Under the QCBS
approach, as envisaged under these laws,
firms are awarded points for price, quality
and preference criteria, and the firm that
scores the highest points gets awarded the
job. However, more emphasis is placed on
price, while quality and functionality are
de-emphasised, as the threshold for quality
is limited to the minimum acceptable level
(CESA 2014).
In response to cost-based selection
methods adopted by clients, consulting
engineering firms have adopted a ‘market
driven’ strategy of price-based competi-
tion and no longer compete based on
quality of services offered (Davies 2006;
Love & Edwards 2004). The rationalisation
for this lowest cost-based selection for
consulting engineering services includes
“… clients wanting to pay fees that conform
to existing market practice, the availability
of qualified firms willing to do work at
lower fees, and the need for cost saving.”
(Ling 2004).
While some countries, such as the
United States of America (USA) and
the United Kingdom (UK), have since
abolished professional fee scales, and
some, such as the USA and Japan, only
use quality-based selection (QBS), in
South Africa guidelines for determining
professional fees are obtainable, and price-
based competition is not prohibited. This
is because the Competition Commission
has outlawed mandatory fee scale
guidelines by professional bodies such as
ECSA, as it is interpreted as price fixing
and therefore in contravention of the
Competition Act. In this regard ECSA has
modified the fees scale guidelines over the
years by doing away with the scale of fees
and replacing it with a system of tables
and charts to enable the estimation of fees
based on a known project type and scope
of work.
South African civil and structural
engineering professionals have, however,
argued that compensation for their ser-
vices as provided for under the ECSA
guideline is inadequate, especially on
building projects, mainly due to the
increased frequency of redesigns and
coordination inputs required on building
projects. Increased construction monitor-
ing and supervision responsibilities are
also implied by the nature of their service,
which has not been adequately reflected
under ‘normal service’.
The consulting engineer’s risks
Risk to the consulting engineer for the
purpose of this research is defined as:
An event, condition or circumstance
before, during or after the execution of a
project that puts the consulting engineer
in an unfavourable position with regard
to meeting clients’ expectation, meeting
professional and ethical standards and
the possibility of incurring financial
liability or reputational damage.
Low professional fees have been observed
to influence conditions and/or attitudes
that create risk for consulting engineers.
This is evident in the findings of a
questionnaire survey of members of the
Association of Consulting Engineers in the
UK (Hoxley 2000). The survey revealed
that, for projects with low fees, consulting
engineers:
■■ give less consideration to design alterna-
tives and to checking and reviewing
of drawings;
■■ consider the risks of design error occur-
ring as being higher;
■■ produce simpler designs to minimise
the commitment of resources to a
design task;
■■ judge the number of claims for addi-
tional fees to be higher on projects with
low fees;
■■ think that low fees engender mistrust
between client and consulting engineer;
and
■■ bid low with the intention of doing less
than in the enquiry, and making up fees
with claims and variation.
The implication of these is a decrease in
the quality of professional services, which
creates risk for the consulting engineer’s
practice. Quality issues often result in
“unsafe structures, delays, cost overruns
and disputes in construction contracts”
(FIDIC 2004). On the one hand, client
dissatisfaction resulting from poor project
performance is not desirable if a consulting
practice is to be successful. On the other
hand, however, quality service on the project
and a satisfied client will, to a great extent,
ensure a repeat appointment from the same
client and also referrals (Jaafar et al 2008).
The risks exposure of engineering ser-
vices consultants is hence two-pronged –
one aspect of their risk exposure is liability
for the professional services they offer to
their clients, while the other aspect con-
cerns threats to running a financially viable
business. These risks are categorised into
project risks and organisational business
risks, and are discussed below.

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 11
Table 1 Summary of consulting engineers’ risk as identified by various authors
Project-specific
Risk factors Reference
risk groups

Type of client and structure of client organisation.

Size of client organisation relative to consulting firm. A client organisation that is significantly larger than the
consulting organisation can cause harm to the consulting firm from a legal perspective.

Depending on a single client for a substantial portion of the firm’s annual fee income.

Non-payment or late payment by client.

IEA 2005
Client insolvency; inability of client to meet financial obligation. Tang et al 2007
Client risk Kometa et al 1996
Unrealistic project duration target set by clients.
Jerling 2009
Poor communication and decision-making structure within client organisation. Maritz & Robertson 2012

Client’s legal history; clients having a reputation of being litigious.

Past experience with client.

Client performance on previous projects; client having a history of unsuccessful projects.

Vulnerability of client to prevailing political and economic situation.

Unclear brief that does not properly state the role and responsibility expected of each party.

Client’s expectations being higher than those communicated to and understood by the engineer.

Scope creep – broadening scope of services.

Unclear procedure for variation and expansion of project scope.

Providing professional advice without a written professional services agreement in place.

Brief, scope and Professional services agreement that does not clearly define terms and conditions of appointment. IEA 2005
contractual risk Victor et al 2014
Unbalanced risk allocation and limits of contractual liabilities.

Terms of contract that are ambiguous and unspecific in terms of:

■■ Scope of normal services
■■ Scope of additional services

■■ Quantum and timing of compensation

■■ Site staff for the project

■■ Channels of communication

■■ Client obligations and responsibilities.

“Time boxing” (allocating set time) of design task.

Inability to perform value engineering.

Low professional fee.

Designs not being reviewed by experienced colleagues.

Design and Bubshait et al 1998
documentation- Time pressure resulting from unrealistic project schedule. Lopez et al 2010
related risk Love et al 2008
Weak or absence of quality control systems in the design process.

Insufficient knowledge, skill and experience within the design organisation.

Over-dependence and ineffective utilisation of computer-aided design programs.

Poor coordination and communication between members of the project team.

Method of project procurement (fast-tracked project procurement methods that compress the schedule
between the design and construction phases of the project or allow both phases to go on concurrently).

Understaffing of project / staff strength and work load of firm. Burati et al 1992
Lopez et al 2010
Rework of design Quality of staff assigned to the project. Love et al 2000
and drawings Schedule pressure. Love 2002
Love & Edwards 2004
Inability of client to make project decisions timely and correctly. Love & Li 2000

Quality and timeliness of information from other project consultants (architect, electrical and mechanical
engineers).

Professional services agreement that limits inspection responsibility or totally redirects such responsibilities
to the client. Kagan et al 1986
Quality control
Tang et al 2003
on site risk Inexperienced site staff engaged for construction inspection services.
Kerkes 2006
Professional negligence with regard to construction inspection.

12 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 Project-specific
Risk factors Reference
risk groups

Conflict of interest; consultants using their positions for financial gain.

Revealing tender information.

Concealing construction faults, poor workmanship and material quality during inspection.
Bowen et al 2007
Unethical Altering of construction documents. Vee & Skitmore 2003
practices Main consultant cutting the fees of other consultants. Kerkes 2006
Ayat 2013
Consultant withholding information from the client which results in variations.

False promises of project advancement.

Misleading clients on project management.

Performing work ‘at risk’.

Below-cost tendering.

Client’s treasury and financial control system (applicable to public clients in South Africa) not suitable for Jerling 2009
Financial risk
project’s financial requirement. IEA 2005

Unreasonable levels of contractual penalties and liabilities.

Contract payment tied to project milestones.

Difficulty affording professional indemnity insurance premiums.

Watermeyer & Smith 2014
Professional Premium loadings being applied to jobs evaluated by insurers to be high risk, e.g. low-fee jobs.
Padayachee 2011
indemnity risk
Most cost-cutting measures adopted by consulting firms border on dishonesty and unethical conduct, IEA 2005
which are grounds for denial of claims by insurers.

Project risks
Project risks are those that threaten the
attainment of project objectives in terms of
cost, time, quality, environmental sustain-
ability and safety. The occurrence of such
risks could have serious consequences
for project stakeholders, including clients
paying more as a result of overruns on
the project budget; reputational dam-
age, indemnity claims and loss of future
appointments for the consulting engineer;
and loss of revenue for the contractor due
to penalties (Visser & Joubert 2008). From
the perspective of the consulting engineer’s
practice, the possible impact of low fees
on these risks is presented under nine risk
groups. Table 1 summarises project risk
factors identified by a variety of authors.
companies operating within the industry
having weak risk management cultures,
frameworks and practices, but also having
high risk management awareness at the
project level (Visser & Joubert 2008). The
emphasis of these companies appear to
be on project level risks and less concern
for the risk at the organisational level.
Visser and Joubert (2008) cited the work
of Von Widden and Black (2007) who
categorised risks to construction business
using the Marsh ‘Risk Universe’. Table 2
presents organisational business risks
and their potential impact as it affects
Table 2 C
 onsulting engineers’ organisational business risks (adapted from Visser & Joubert (2008))
No Organisational business risks
1 Shortage of key skills (human capital)
consulting engineering firms based on the
risk framework developed by Visser and
Joubert (2008).
Failure to adapt to rapid changes in
the construction industry is the biggest
threat to the business of AEC (architecture,
engineering and construction) firms;
firms that are unable to adapt to these
changes will become marginalised and fail
(Gupta 2012). The increased complexity
and sophistication of construction projects,
rapid technological changes, changes in
project delivery methods and globalisation
have transformed design processes and the
Potential impact
Poor workmanship

Organisational business risks 2 Tendering and contract exposures Legal exposures

The failure of construction projects concur-
Identification, reporting and actioning of Project management and quality control
rently managed by a consulting engineering 3
project non-conformance issues
firm at the same time may not only be as a
4 Poor business risk management Guarantee exposure/business sustenance
result of failure of the firm to manage risk at
the project level, but also at the organisation- 5 Project management issues Delays and penalties
al level (Liu et al 2013). Organisational risks 6 Poor data management Operational exposure
can be considered as threats to the business
Financial fluctuations and cost overruns on
and operational activities of the organisation, 7 Financial/cost exposure
long-term projects
and impact on the competitiveness of the
organisation. These risks, if not controlled, 8 Government and legislation issues Curtailed options/legal exposure

could result in economic losses for the firm. 9 Client relationships Repeat job/business sustenance
The construction business environ-
10 Dearth of innovation Competitiveness/ reputational issues
ment in South Africa is full of risks, with

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 13
way engineers collaborate. Globalisation
has removed geographic constraints, and
has resulted in AEC firms being able to
provide their services across international
borders. Technology has transformed the
way information is shared and accessed,
while the increasing complexity of
engineering projects is driving greater
specialisation within the various engineer-
ing disciplines. The implication of these for
AEC firms is increased competition, as well
as opportunities.
The sustenance of a consulting engineer-
ing practice depends on the firm’s ability
to develop strategies to improve competi-
tiveness and secure adequate volumes of
projects that satisfy cash flow requirements
and profitability (Jaafar et al 2008).
THE RESEARCH CONDUCTED
Given the exploratory nature of the study,
a survey research method was applied.
A structured questionnaire was used as
the data collection tool for this research.
The questionnaire aimed to address
the research objectives by investigating
the following:
a. Project and organisational risks encoun-
tered by consulting engineers.
b. The extent of the practice of discount-
ing amongst consulting engineers.
c. The impact of discounted fees on pro-
ject and organisational risk factors by
frequency of occurrence and magnitude
of impact.
The questionnaire contained both open
and close-ended questions to allow the
researcher to analyse the responses both
qualitatively and quantitatively. A web-
based tool was used for the design and
distribution of the survey questionnaire.
The questionnaire consisted of three
sections: personal information (­section 1),
organisation’s discounting practice
(­section 2) and risk factors (section 3).
The objectives of questions in section 1
included categorisation of firms, identifica-
tion of trends associated with organisations
with similar characteristics, identifying
respondents from the same organisation,
and determining the level of experience
and participation of the respondent in the
organisation’s decision-making. Questions
in section 2 aimed to understand the
practice of discounting among consulting
engineering firms, and factors that influence
the practice. In section 3, questions about
project and organisation risks peculiar to
consulting engineering firms were asked.
14
The insights gained from the review of
literature on the subject of risk particular to
consulting engineers formed the basis for
questions posed to the industry in this sec-
tion of the questionnaire. Respondents were
asked to rate the significance of discounted
fees on identified project and organisational
risk in terms of frequency of occurrence and
impact. Questions were also asked to iden-
tify areas of professional services that are
likely to be compromised when professional
fees are low, and risk management practices
employed to mitigate the impact of low fees.
Survey sample and selection
of respondents
The target respondents for this research were
ECSA-registered civil and structural profes-
sional engineers practising in South Africa.
A purposive sampling technique was used
in the selection of the survey sample for this
research. The purposive sampling technique
has been identified to be suitable for quan-
titative and qualitative research methods
(Tongco 2007). Emuze and Smallwood (2013)
argue that the purposive sampling technique
can be used whenever the characteristics of
the population cannot be precisely deter-
mined, and in such instances the research
sample can be made up of informants who
the researcher believes to be a representative
of the population under investigation.
Respondents for the survey were
selected from two sources. The first was
a working group of structural engineering
consultants who in 2013 drafted a docu-
ment to push for a review of the published
fee scale for structural engineering services
on building projects under the ‘ECSA
Guideline for Defining the Scope of Service
and Determining Professional Fees Scales’.
Eight members of that working group were
identified as respondents for this research.
The second source was a database of civil
and structural engineers who had partici-
pated in continuing professional develop-
ment (CPD) courses at the civil engineering
department of Stellenbosch University.
Participants at these courses are middle-level
and senior-level employees at both construc-
tion and consulting engineering firms. This
database was consulted to identify suitably
experienced practising consulting engineers.
A total of 61 consulting engineers were
considered suitable for selection as respond-
ents, based on their years of experience,
area of competence and position within
the organisation. The questionnaire was
then emailed to them. Of this number, 23
responses were received, representing a 38%
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Table 3 Likert scale conversion – Method 1
Conversion/
Frequency scale
interpretation
Never Not common
Seldom Not common
Often Common
Frequently Common
Table 4 Likert scale conversion – Method 2
Conversion/
Agreement scale
interpretation
Strongly disagree No
Disagree No
Neither agree nor disagree Undecided
Agree Yes
Strongly agree Yes
response rate. This response rate is consid-
ered satisfactory, considering that industry
participants in South Africa have a history
of reluctance to participate in questionnaire
surveys distributed by mail or other elec-
tronic means (Ugwu & Haupt 2007).
Method of data analysis employed
Analysis of data received from the ques-
tionnaire survey was carried out using
frequency analysis and ranking. Most
questions required answers to be provided
by means of a Likert type scale. Three
variants of the scale were used. The first
was aimed at determining the frequency of
occurrence, the second was to measure the
extent to which respondents agreed with
certain observations, and the third aimed
at measuring impact. The method of com-
bination and conversion of these scales for
the purpose of qualitative and quantitative
analysis is presented in Tables 3–5.
ANALYSIS AND SYNTHESIS
OF SURVEY FINDINGS
Personal information
Analysis of personal information shows
that the majority (65%) of respondents were
active in the structures and building market,
followed by the civil works market. They
represented small, medium and large-sized
consulting engineering firms. The personal
information also defines the majority (75%)
of respondents to be involved in decision-
making and management roles within their
organisations, and as such their responses
can be taken to be indicative of their Table 5 Likert scale conversion – Method 3
organisation’s position on the subject being Weighting
investigated. Sixty one percent of respond- Frequency and impact scale
Non-linear scale Linear scale
ents have been practising for over 15 years
and have spent an almost equal number Frequently 1.0 1.00
of years with their current employer. It is Often 0.8 0.66
observed that the years of experience in the Frequency
industry closely compare with the number Seldom 0.3 0.33

of years most respondents had been with Never 0 0

their current employer. This is indicative of
Severe 1.0 1.0
the level of staff mobility within the sampled
respondents’ organisations. Significant 0.8 0.66
Impact
Minimal 0.3 0.33
Discounting practices in
No impact 0 0
respondents’ organisations

Type of clients and method Table 6 Methods of remuneration

of remuneration Commonly used
Respondents were asked the type of clients Method of remuneration
Count Percentage
who accounted for the bulk of their fee
earning. It was observed that firms with Percentage fees based on the cost of the works 17 77.3%
less than ten employees all had private
Time-based fees 15 68.2%
clients accounting for up to 90% of their
fee earnings. The two firms that had public Reimbursable expenses 12 54.6%
clients accounting for up to 90% of their fee Fixed sum 10 45.5%
earnings had similar profiles in terms of
Value-based fees 5 22.7%
number of employees (50 to 99) and annual
turnover of between ZAR 10 million and
ZAR 100 million. Most of the firms with Table 7 Reason for offering discounts
an annual turnover above ZAR 250 million Yes No Undecided
serviced both public and private clients. Primary reasons
The methods of remuneration common Count % Count % Count %

to consulting engineering organisations Secure new clients (client relations) 18 81.8% 0 0% 4 18.2%
was established. Table 6 presents the data
Test a new market 7 31.8% 1 4.6% 14 63.6%
received from respondents. Percentage
fees based on the cost of the works were To keep staff busy and cover running cost 12 54.6% 3 13.6% 7 31.8%
found to be the most common method of
Forced by prevailing market conditions 21 90.9% 1 4.6% 1 4.6%
remuneration among respondents’ organi-
sations. This is followed by time-based
fees, reimbursable expenses and fixed sum. had this opinion: “I run a cost-efficient small of a structured question. Table 7 presents
Value-based fees are the least commonly practice. I don’t offer discounts, I just charge the responses received.
used methods of remuneration. less than I should, because the profession An analysis of the responses suggests
has historically been abused by the client that prevailing market conditions and cli-
Prevalence of discounting body. In my opinion this situation evolved ents’ demands for discounts were the main
The prevalence of discounted fees bench- because we, the professionals, allowed it.” reasons why consulting engineers offer
marked against the ECSA recommended discounts on fees.
fees scale was assessed. Respondents Reason for offering discounts This trend is further supported by
were asked if their organisations offered The primary reason why discounts are responses to an open-ended question
discounts on the ECSA recommended fees offered in the industry was assessed by way which allowed respondents to provide
scale during tender or negotiation with
clients. Figure 1 presents the responses
received, which suggest that the practice Frequently 56.52%
of discounting is quite prevalent, as 83% of
Often 26.09%
respondents either frequently or often offer
Seldom 8.70%
discounts on fees during tender or during
negotiations with clients. No response 4.55%
Responses to an open-ended question 0 10 20 30 40 50 60 70
revealed that, even though some engineers Percentage (%)
do not explicitly offer discounts, they
however reduced their fees. One respondent Figure 1 Prevalence of discounting practice among respondents

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 15
additional information on the reason they
offer discounts. A total of nine respondents No response
completed the open-ended questions. The 46–60%
1–15%
majority (56%) of the comments referred
to pressure from clients to offer discounts. 9%
31–45% 5%
Three respondents identified non-regulation 23%
of the different civil engineering disciplines 5%
as worsening the impact of competitive
tendering on the pricing of consulting engi-
neering services. Some of the responses that
point to client factors include:
“I don’t like the concept of offering
discounts, but clients often insist on 59%
discounted fees.”
“For private clients, discounts are
expected. Many private clients even
expect free work (work done at risk) dur-
ing the feasibility stage, with a promise 16–30%
to appoint the consultant to the project
should it go ahead.” Figure 2 Average percentage of discount offered on fees
“In some markets the fees are simply
stated by the client / project manager /
QS, and if you want the job you must No response
1–15%
accept a certain percentage fee. This is
Above 60%
common on commercial projects such as
5% 9%
shopping centres.”
9%
Some other respondents argue that the
market is unregulated, and as such the
discounting practices are influenced by
prevailing market conditions.

Amount of discount offered
Respondents were asked about the aver-
age and maximum amounts of discount
they offered as a percentage of their fees.
Responses received are presented in
Figure 2 and Figure 3.
Analysis of the data shows that the
majority of the respondents offered an
average discount of between 16% and 30%
on the ECSA-approved fees scale. A sub-
stantial proportion of respondents reported
that they have offered discounts of between
46% and 60% on the ECSA-approved fees
scale. No trend was observed regarding
the amount of discount offered by small,
medium and large firms, as no range of
discount was particular to any category of
firm irrespective of size, the market seg-
ment they operated in or the type of client
they serviced.
Impact of discounted fees on risks
to the consulting engineer’s practice
Analysis of impact of discounted
fees on project specific risks
Project-specific risks particular to consult-
ing engineers were identified from the
review of literature conducted, and the
27%
46–60%
9%
31–45%
Figure 3 Maximum percentage of discount offered on fees
influence of low professional fees on these
risks quantitatively assessed. Their fre-
quency of occurrence on projects with low
professional fees, and their impact on the
consulting engineer’s practice were assessed
using a Likert-type scale (see Table 5). The
objective was to rank the various project
risks in an order of perceived importance by
obtaining risk values for each project risk
considered. In order to obtain risk values the
different response options were weighted
using a linear and non-linear weighting
scale. A non-linear weighting scale was
introduced to accentuate the significance of
frequently occurring risk with high impact;
this approach is consistent with the method-
ology employed by Jerling (2009) in ranking
the risk generated for contractors by clients.
41%
16–30%
The risk value of each risk group was
obtained using the risk expression:
Risk = Risk Frequency × Risk Impact
In order to compare the risk, the weighted
value of the response option was multiplied
by the number of responses for that option.
A final risk value was obtained using the
expression:
Risk Value = [(N Number of responses ×
Frequency Weight) ×
(N Number of responses ×
Impact Weight)]
Project risks with higher risk values are
considered to be more influenced by low

16 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Table 8 Ranking of project risk groups

Risk description (linear scale) Rank Risk description (non-linear scale)

Financial loss on the project 1 1 Financial loss on the project

Inability to perform value engineering 2 2 Inability to perform value engineering

Inadequate site supervision and quality control 3 3 Inadequate supervision and quality control

Rework of design and drawings 4 4 Rework of design and drawings

Poor quality of design and documentation 5 5 Poor quality of design and documentation

Legal liabilities (claims, disputes and litigation) 6 6 Legal liabilities (claims, disputes and litigation)

Professional indemnity cover 7 7 Professional indemnity cover

Unethical practices 8 8 Unethical practices

professional fees and are ranked above
project risks with lower risk values. The
ranking of project risks based on the risk
values obtained using both the linear and
non-linear weighting scales is presented in
Table 8.
Financial loss on the project, inability
to perform value engineering, inadequate
supervision and quality control issues,
rework of design and drawings, and poor
quality of design and documentation are
identified as the top five project risk groups
impacted by low professional fees.
Impact of discounted fees on
organisational business risks
Seven organisational risks were
identified as being influenced by low
professional fees. These seven organisa-
tional business risks were assessed in the
questionnaire survey.
Organisational business risks are
assumed to always be present in the
organisation’s day-to-day running. The
managers of the business therefore need
to constantly steer the organisation away
from these threats. In this light, assessing
these risks in terms of frequency of occur-
rence was not feasible; respondents were
therefore only asked to assess the impact of
discounted professional fees on the identi-
fied organisational risk factors.
The impact of low fees on these identi-
fied risk factors was assessed quantitatively
and ranked in an order indicative of the
extent to which they are impacted by low
fees. The linear and non-linear weighting
scales previously employed and shown
in Table 5 were also applied. A risk value
was obtained for each risk factor using
the expression:
Risk Value = [ N Number of responses ×
Impact Weight]
Table 9 presents the results of the ranking
process. Human resource issues (training
and mentoring of young engineers, and
attracting experienced engineers) appear
to be impacted most by discounted fees,
followed by business sustenance and
technical innovation within the organi-
sation. Staff morale, relationship with
clients and the organisation’s corporate
reputation were ranked the least impacted
organisational risks.
Two open-ended questions were also
included in the questionnaire for the
purpose of obtaining additional informa-
tion on project and organisational risks
resulting from discounted professional fees.
Respondents were asked to provide their
perspective on the impact of discounted
fees on the project risk and organisational
risk exposure of consulting engineers. A
total of 22 respondents provided answers to
these questions.
The responses provided were not at
variance with the project risks and organi-
sational risks captured in the structured
questions. Analysis of these responses was
carried out qualitatively by interpreting
respondents’ opinions and categoris-
ing these into the identified common
risk groups.
The majority of the opinions highlight-
ed the limited resources and time deployed
to projects, with low fees resulting in risks
related to quality of design documents.
Other risks that received considerable men-
tion include business sustenance, ability to
train young engineers and to retain the ser-
vices of experienced engineers, all factors
critical to the sustenance of the profession.
Devaluation of the profession, relationship
between engineer and client, increased
workload for consulting engineers and

Table 9 Ranking of organisational business risks

Risk description (linear scale) Rank Risk description (non-linear scale)

Ability to train and mentor aspirant/graduate engineers 1 1 Ability to train and mentor aspirant/graduate engineers

Ability of organisation to attract and retain quality/experienced staff 2 2 Ability of organisation to attract and retain quality/experienced staff

Business sustenance 3 3 Business sustenance

Ability of organisation to be innovative in design 4 4 Ability of organisation to be innovative in design

Motivation of staff 5 5 Motivation of staff

Relationship with clients 6 6 Relationship with clients

Reputation of your organisation 7 7 Reputation of your organisation

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 17
demotivation of engineering staff were
also mentioned. Analysis and collation of
the responses received are presented in
Table 10.
Strategies employed for the
management of risks associated
with discounted fees
In dealing with risk associated with dis-
counted fees, respondents either contractu-
ally limited their risk exposure, or mostly
accepted the risks and employed various
mitigation strategies, including:
a. Avoiding innovative ideas in order to
cut down on time spent on the project.
b. Accepting financial loss on the project,
but guarding against technical risks by
dedicating adequate time and engineer-
ing input to the project.
c. Negotiating reduced scope of service to
accompany discounted fee.
d. Recovering cost on future projects with
the specific client.
e. Charging inflated prices for variations
and refusing to do work not included in
the scope.
f. Doing only jobs which are repetitive at
low fees so that information can be used
from previous projects.
g. Improving organisational efficiency
through the use of technology.
h. Adopting a quality assurance system.
DISCUSSION AND
IMPLICATION OF FINDINGS
Findings from the survey show that
discounting practices are widespread
among the structural and civil engineer-
ing services consulting firms represented
in this survey. At least seven in every ten
respondents reported that they either
frequently or often offered discounts.
Discounts of between 16 and 30% of the
recommended fee scale are observed to be
the industry’s standard.
The majority of the respondent firms
can be described as medium-sized firms
judging by their staff strength and annual
turnover. No clear dichotomy was, how-
ever, observed between the discounting
practices of small, medium and large firms
represented in the survey.
Percentage fees based on the cost of the
works is by far the most common method
of remuneration amongst respondent
organisations, followed by time-based fees.
The monetary value of remuneration under
both these methods is informed by rates
recommended in the fee scale guidelines
18
Table 10 Risk identified by respondents in the open-ended question
Risk associated with discounted fees identified by respondents
Marginal profit or financial loss on projects
Decreased quality of service/inability to perform value engineering
Ability to hire and retain highly qualified staff
Staff demotivation and fatigue resulting from increased workload on engineers
Design errors
Business sustenance (sustainability of the practice)
Poor quality design and documentation
Inability to afford staff training
Legal problems
Inability to afford competitive salaries
Client relationship suffers
Devaluation/deterioration of the profession
Professional indemnity claims
Absence of innovation in design
Inability to carry out proper quality control on site
Huge financial implication of redesigns can erode profit
published by ECSA. This further suggests
that the ECSA fee scale is widely used in
the industry.
Competition based on price appears to
be a well-entrenched business strategy as
over two thirds of respondents agreed that
their firm competed based on price. This
finding agrees with the assertion by Love
and Edwards (2004), that consulting firms
are increasingly adopting a ‘market driven’
strategy of price-based competition and no
longer compete based on quality of service.
The study also found that prevail-
ing market conditions characterised
by competitive tendering and demands
from clients for discounts are the leading
reasons why respondents offer discounts.
Responses to an open-ended question
reveal that, because the ECSA fees scale
is not enforceable and the practice of
professional engineers within the various
engineering discipline is unregulated,
clients have a greater bargaining leverage
when it comes to negotiating fees with the
consulting engineer.
The level of discounts offered in the
industry, evident in the findings of this
research, calls to question the relevance of
the ECSA fees scales. Some respondents
argued in the open-ended questions that
the fees scales are no longer relevant and
that, rather than use fee scales, engineers
should arrive at their own fees through a
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Number of
times recorded
11
9
8
8
6
6
5
5
4
4
4
3
2
1
1
1
careful consideration of the scope of the
service they are to provide. This argument
hinges on the fact that much of the fees
determined in the industry are based on a
percentage of the cost of work, but current
realities, especially on building projects,
show that ECSA-recommended percentage
scales may not be an accurate reflection of
the engineering input required from the
consulting engineer.
It is found that the quality of profession-
al service offered is impacted by discounted
fees, especially as consulting firms do not
spend enough time considering design
options and carrying out value engineering
on low-fee jobs. The implication of this is
that the life cycle cost of projects increases
on account of poor engineering inputs
(Sterner 2000). When this becomes appar-
ent to clients, they may no longer value the
services of the engineering professional,
leading to more design responsibilities
being awarded to contractors in the form of
design-build contracts.
The construction industry is reputed
to be plagued by adversarial relationships
between project participants leading to
conflicts (Black et al 2000). Findings here
have shown that discounted fees negatively
impact on the relationship between the
consulting engineer and the client. This
is a disincentive for consulting engineers
to embrace partnering philosophies and
collaboration on projects, as they will
naturally spend less time and deploy fewer
resources on projects with low fees.
These findings strongly indicate that
government’s procurement legislations,
including those of the Competition
Commission, that encourage competitive
procurement based on cost have engen-
dered price-based competition in the
consulting engineering industry. Two key
implications for the industry and the coun-
try as a whole are discussed below:
a. Increase in the life cycle cost of
projects: Although the cost of profes-
sional engineering services on projects
constitutes a very small portion of the
life cycle cost of engineering projects
(Shrestha & Mani 2013), requiring
consulting engineers to compete based
on cost rather than quality to secure
contracts inherently requires that the
amount of engineering input (by way
of considering design alternatives and
carrying value engineering) on the
project is reduced to enable the engi-
neer to bid low and still make a profit.
This can lead to increased operation
and maintenance cost in the long run,
and hence increased life cycle cost of
the project.
b. Shortage of professionals: When com-
pany profits are low, remuneration for
consulting engineers will be affected. To
maximise profit, individual personnel
within the firm will hence have to take
on responsibilities not commensurate
with remuneration, which otherwise
could have been delegated to other
personnel. Similarly, the ECSA remu-
neration scale makes certain types of
projects (e.g. building projects) in cer-
tain markets less profitable than others.
If consulting engineers operating in that
market are not protected, a shortage of
professionals on such projects may be
recorded.
These ultimately make the profession
unattractive to young engineers, and could
lead to capital flight of existing engineers
and lack of enrolment of younger minds
in engineering training. As such, fewer
engineers are available to design and imple-
ment engineering projects required for
national development.
RECOMMENDATIONS AND
FUTURE RESEARCH
Based on the findings of this study, the fol-
lowing recommendations are suggested:
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
a. The ECSA fees scale tariff has become
a tool in the hands of clients to exploit
consulting engineers, as it is neither
enforceable nor mandatory. Its relevance
in the face of the current procurement
legislation under which it is implement-
ed hence needs to be questioned.
b. In countries such as the USA and Japan,
which are recognised as technologically
advanced, quality-based selection is
law for the procurement of consulting
engineering services. It is recommended
that procurement legislation for profes-
sional engineering services be amended
to allow the selection of consulting
engineers based solely on qualification,
experience and competence. Fees can
subsequently be determined based
on the cost of providing the service
required under a well-defined scope of
work
c. Registered professionals who do not
possess the correct experience and
training have often been accused of
offering reckless amounts of discounts
because they are sometimes unaware
of the amount of engineering input
required and the level of risk involved.
Civil engineering is a broad field of
engineering that encompasses many
sub-disciplines. Regulating the practice
requirements within the various sub-
disciplines in the field of civil engineer-
ing, such as structural engineering,
is recommended. It is believed that
properly qualified professionals will be
unwilling to offer ridiculous discounts.
d. Further research on the implication in
South Africa of lower quality service
by the consulting engineer, engendered
by discounted fees, on the risk encoun-
tered by other project participants
such as clients and contractors, is also
recommended.
CONCLUSION
Research on construction industry risks
have focused much on client and contrac-
tor risks. The research conducted in this
study investigated risks to the consulting
engineer’s practice, with focus on the
impact of discounted professional fees on
the consultant’s risks.
Using survey questionnaires, the study
yielded information on the local experience
of consulting engineers in South Africa,
and ranked project and organisational risks
encountered by consulting engineers based
on how they are perceived to be impacted
by low fees. A key finding of this study
suggests that discounted professional fees
accentuate the risk exposure of structural
and civil engineering consulting profes-
sionals. This is consistent with widely
held perceptions amongst structural and
civil engineering professionals within
the South African industry. The study
hence confirms anecdotal expectations,
using empirical methods, and provides
an academic condensation of industry-
wide practices and challenges concerning
consulting engineering practice. These
findings are relevant, not only for industry
consulting professionals, but also for other
stakeholders in the construction industry,
particularly policy makers, client bodies
and regulators. The significance of this
study furthermore lies in the fact that, to
the best of the authors’ knowledge, this
study provides the first known empirical
and academic work for the South African
construction consulting environment,
and even for Africa and the Third World,
offering relevant information represent-
ing the realities in these less-researched
environments.
REFERENCES
Ayat, YA 2013. Unethical conduct among professionals
in construction industry. Unpublished Master’s
dissertation. Gaza: The Islamic University of Gaza.
Black, C, Akintoye, A & Fitzgerald, E 2000. An analysis
of success factors and benefits of partnering in
construction. International Journal of Project
Management, 18(6): 423–434.
Bowen, P, Akintoye, A, Pearl, R & Edwards, PJ 2007.
Ethical behaviour in the South African construction
industry. Construction Management and Economics,
25(6): 631–648.
Bubshait, A A, Al-Said, FA & Abolnour, M M 1998.
Design fee versus design deficiency. Journal of
Architectural Engineering, 4(2): 44–46.
Burati, J, Farrington, J & Ledbetter, W 1992. Causes
of quality deviations in design and construction.
Journal of Construction Engineering and
Management, 118(1): 34–49.
CESA (Consulting Engineers South Africa) 2007. State
of the South African Consulting Profession in South
Africa – January to June 2007. Johannesburg: CESA
CESA (Consulting Engineers South Africa) 2013.
Bi-Annual Economic and Capacity Survey – January
to June 2013. Johannesburg: CESA.
CESA (Consulting Engineers South Africa)
2014. Best practice: Procurement of consulting
engineering services in South Africa (discussion
paper). Available at: http://www.cesa.co.za/
sites/default/files/20140410_CESA%20
Position%20Paper_Best%20Practice%20
19
 Procurement_Consulting%20Engineering%20
Services.pdf (accessed on 17 September 2014).
Davies, B 2006. A review of South African and
international practices for the procurement and
remuneration of consulting engineering services.
Unpublished Master’s dissertation. Stellenbosch:
Stellenbosch University. Available at: http://hdl.
handle.net/10019.1/50577.
Emuze, F & Smallwood, J 2013. Management concepts
and project performance: Perceptions from the
South African public sector environment. Journal
of the South African Institution of Civil Engineering,
55(2): 21–28.
FIDIC (International Federation of Consulting
Engineers) 2004. Quality of construction. FIDIC
policy statement. Available at: http://www.fidic.
org/sites/default/files/fidic_policy_quality.pdf
(accessed on 17 June 2014).
Gupta, R P 2012. Embrace change or risk failure:
Engineering firms that avoid change and cling to the
status quo are destined to fail. Consulting Specifying
Engineer, 4964.
Hoxley, M 2000. Are competitive fee tendering and
construction professional service quality mutually
exclusive? Construction Management & Economics,
18(5): 599–605.
IEA (Institution of Engineers Australia) 2005. Risk
Management Strategies Guide. Canberra, Australia:
IEA.
Jaafar, M, Aziz, A R A & Wai, A L S 2008. Marketing
practices of professional engineering consulting
firms: Implement or not to implement? Journal of
Civil Engineering and Management, 14(3): 199–206.
Jerling, W 2009. Construction risks generated by
employers from the perspective of South African
contractors. Unpublished MSc dissertation.
Bath, UK: Department of Architecture and Civil
Engineering, University of Bath.
Kagan, H, Leary, D & Pratter, G 1986. Design
engineers’ responsibilities during construction.
Journal of Construction Engineering and
Management, 112(3): 394–402.
Kerkes, D J 2006. Why engineers get sued and how to
reduce your risk. Paper presented at the ASCE Texas
Section Spring Meeting, Beaumont, TX.
Kometa, ST, Olomolaiye, P O & Harris, FC 1996.
A review of client-generated risks to project
20
consultants. International Journal of Project
Management, 14(5): 273–279.
Ling, F Y Y 2004. Consultancy fees: Dichotomy
between A/E’s need to maximize profit and
employers’ need to minimize cost. Journal of
Professional Issues in Engineering Education and
Practice, 130(2): 120–123.
Liu, J Y, Zou, P X & Gong, W 2013. Managing project
risk at the enterprise level: Exploratory case studies
in China. Journal of Construction Engineering and
Management, 139(9): 1268–1274.
Lopez, R, Love, P E, Edwards, D J & Davis, PR
2010. Design error classification, causation, and
prevention in construction engineering. Journal
of Performance of Constructed Facilities, 24(4):
399–408.
Love, P & Li, H 2000. Overcoming the problems
associated with quality certification. Construction
Management & Economics, 18(2): 139–49.
Love, P E 2002. Influence of project type and
procurement method on rework costs in building
construction projects. Journal of Construction
Engineering and Management, 128(1): 18–29.
Love, P E & Edwards, D J 2004. Forensic project
management: The underlying causes of rework
in construction projects. Civil Engineering and
Environmental Systems, 21(3): 207–228.
Love, P E, Edwards, D J & Irani, Z 2008. Forensic
project management: An exploratory examination
of the causal behavior of design-induced rework.
IEEE Transactions on Engineering Management,
55(2): 234–247.
Love, P E, Mandal, P, Smith, J, Irani, Z, Treloar, G
& Faniran, O 2000. DECOREM: A design and
construction rework minimisation model. Paper
presented at the 1st International Conference
on Systems Thinking in Management (ICSTM),
8–10 November, Geelong, Australia.
Lu, W, Ye, K, Flanagan, R & Jewell, C 2013.
Developing construction professional services
in the international market: SWOT analysis of
China. Journal of Management in Engineering,
29(3): 302–313.
Maritz, M & Robertson, D 2012. What are the legal
remedies available to contractors and consultants
to enforce payment? Journal of the South African
Institution of Civil Engineering, 54(2): 27–35.
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Padayachee, M 2011. Report of the AON Professional
Indemnity Insurance Scheme in the CESA Year End
Report 2011. Johannesburg: CESA.
Shrestha, P P & Mani, N 2013. Impact of design cost
on project performance of design-bid-build road
projects. Journal of Management in Engineering,
30(3): 04014007.
Sterner, E 2000. Life-cycle costing and its use in the
Swedish building sector. Building Research &
Information, 28(5–6): 387–393.
Tang, S, Lu, M & Chan, Y 2003. Achieving client
satisfaction for engineering consulting firms.
Journal of Management in Engineering, 19(4):
166–172.
Tang, W, Qiang, M, Duffield, C F, Young, D M & Lu, Y
2007. Risk management in the Chinese construction
industry. Journal of Construction Engineering and
Management, 133(12): 944–956.
Thela, Abe 2014. Consulting engineers – not a
commodity but trusted advisor. Available at:
http://www.cesa.co.za/node/421 (accessed on
4 November 2014).
Tongco, M D C 2007. Purposive sampling as a tool for
informant selection. Journal for Plant, People and
Applied Research, 5: 147–158.
Ugwu, O & Haupt, T 2007. Key Performance Indicators
and assessment methods for infrastructure
sustainability. A South African construction
industry perspective. Building and Environment,
42(2): 665–680.
Vee, C & Skitmore, C 2003. Professional Ethics in the
Construction Industry. Engineering, Construction
and Architectural Management, 10(2): 117–27.
Victor, O & Schinnerer & Company Inc. 2014.
Concepts in risk management. Available at: http://
www.schinnerer.com/risk-mgmt/Documents/
UnprotectedFiles/AIA-Membe‑rs/vep1-1.pdf
(accessed on 22 July 2014).
Visser, K & Joubert, P 2008. Risk assessment modelling
for the South African construction industry. Paper
presented at the Portland International Centre
for Management of Engineering & Technology
(PICMET 2008), Cape Town.
Watermeyer, R & Smith, T 2014. Managing structural
engineering risk. Civil Engineering, 22(2): 35–39.
Weidemann, W. 2014. On discounting and tendering.
Civil Engineering, 22(2): 84.
Evaluation of locally TECHNICAL PAPER
available synthetic macro Journal of the South African
Institution of Civil Engineering

fibres in a single-fibre
ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 21–30, Paper 1744

pullout test in concrete COURTNEY ODENDAAL, who is an Associate Member of

SAICE, is a structural engineer at Moroff and Kühne
Consulting Engineers, where she is involved in the
structural design of multi-storey buildings. Prior to this
she worked at Aurecon for two and a half years in a
C M Odendaal, A J Babafemi, R Combrinck, W I de Villiers, W P Boshoff similar capacity. She obtained an MEng in Civil
Engineering from Stellenbosch University in 2014, with research focusing on
performance-based specifications for locally available macro synthetic fibres.
Contact details
There is increasing use of synthetic macro fibres in concrete applications. However, the Moroff and Kuhne Consulting Engineers
Suite 307, The Regent, 19–33 Regent Road, Sea Point
performance of these fibres varies, based on various factors. This study investigates the fibre/ Cape Town 8005, South Africa
matrix interface performance of four types of synthetic macro fibres locally available in the T: +27 21 434 8176, E: courtney.odendaal1@gmail.com

South African construction market. The influence of water-cement ratio (w/c), fibre embedment DR A JOHN BABAFEMI obtained his PhD from the
length, fibre properties and fibre snubbing angle on the single-fibre pull-out behaviour is Department of Civil Engineering, Stellenbosch
University, in March 2015, and was a postdoctoral
studied in this paper. The influence of some of these properties on the compressive strength research fellow at the Unit for Construction Materials in
is also evaluated. Three mix designs and four fibre types were used for the investigation. All the same Department between 2016 and 2017. He is
currently lecturing at the Department of Building at the
tests were performed in a controlled climate room. Specimens with single fibres were tested Obafemi Awolowo University, Ile-Ife, Nigeria. His research interest is in the areas
of fibre-reinforced concrete, concrete durability and use of supplementary
in a Universal Testing Machine. Scanning Electron Microscope (SEM) images of pulled out cement materials. He has several research publications to his credit.
fibres were examined for inferences. The results have shown that w/c has no influence on the Contact details
pull-out load of synthetic macro fibres. However, the pull-out load increases with increase in Unit for Construction Materials, Department of Civil Engineering,
Stellenbosch University, Stellenbosch 7602
fibre embedment length and equivalent fibre diameter. Pull-out load increases for flat fibres T: +27 21 808 4498, E: ajbabafemi@oauife.edu.ng
bent along their strong axis as the fibre snubbing angle increases. While crimped fibres with DR RIAAN COMBRINCK is a senior lecturer at Stellenbosch
improved geometry show optimum pull-out performance when pulled out perpendicularly to University, where he also obtained both his MSc and PhD
on the cracking of plastic concrete in slab-like elements.
the crack plane, flat fibres performed better at the snubbing angles tested. SEM images also He is part of the Unit of Construction Materials which
show more surface damage on fibres with irregular geometry. concentrates on teaching, research and consulting, and
also the head of the structural laboratory, both within
the Civil Engineering Department. He has been involved in various commercial
and academic research projects, and has supervised several postgraduate
students. He also authored or co-authored several journal and conference
INTRODUCTION synthetic fibres (Buratti et al 2011). Areas publications during the past seven years.
Concrete is widely used in the construction of application include concrete ground slabs Contact details
Unit for Construction Materials, Department of Civil Engineering
industry. Due to the inherent weakness (Alani & Beckett 2013; Sorelli et al 2006), Stellenbosch University, Stellenbosch 7602
T: +27 21 808 4946, E: rcom@sun.ac.za
in the tensile properties of concrete, it is concrete pipes (de la Fuente et al 2013),
conventionally reinforced with steel bars. precast structural elements (Ferrara & Meda WIBKE DE VILLIERS is a lecturer at Stellenbosch
University, where she also obtained her MSc on web
More recently the use of short discrete 2006) and shotcreting (Bernard 2004). compression elements in parallel chord trusses. She is
fibres in reinforcing concrete has increased Lately the development and use of larger currently working towards a PhD on the development
of performance-based specifications for alternative
significantly. However, it is still a matter of monofilament synthetic macro fibres (also masonry units. She is part of the Unit of Construction
debate if fibres can completely replace steel called structural fibres) have been on the Materials which concentrates on teaching, research and consulting within the
Civil Engineering Department. She has supervised a number of postgraduate
reinforcement in concrete. The functions increase, which is an indication of their students and authored or co-authored several publications during the past
seven years. Prior to this she spent two years at Aurecon as a structural
of fibres in concrete vary, depending on the viability as an alternative to steel fibres and design engineer.
type (steel, synthetic, etc) and size (micro steel reinforcing bars. They have better Contact details
or macro). While they mitigate shrinkage resistance to chemicals (Wang et al 1987), Unit for Construction Materials, Department of Civil Engineering
Stellenbosch University, Stellenbosch 7602
in some cases, their main function is in the do not deteriorate with age (Hannant T: +27 21 808 4072, E: wdv@sun.ac.za
control of crack opening, thereby enhancing 1998), and are cheaper and lighter
residual tensile properties and durability in (Richardson 2005). However, the mechani- PROF BILLY BOSHOFF is head of the Unit for
Construction Materials at Stellenbosch University. He
the cracked phase of cement composites. cal performance of synthetic macro fibres was also head of the Structural Engineering Division
Due to the bridging mechanism of in concrete could significantly vary due for seven years and served as the president of the
Concrete Society of Southern Africa from 2012 to 2014.
fibres in cracked cementitious composites, to a number of factors, such as geometry He completed his PhD at Stellenbosch University in the
(straight, flat or crimped), diameter, length, field of fibre-reinforced concrete in 2007 and has published more than 78
significant improvement in properties, such papers in journals and conference proceedings. His research interest is in the
as toughness (Buratti et al 2011), fatigue tensile strength, etc. The effect of tempera- field of construction materials, but more specifically fibre-reinforced
concrete, concrete cracking in the fresh state, as well as unconventional and
resistance (Buratti et al 2011), impact ture on the properties of synthetic fibres eco-friendly construction materials.
resistance (Mindess & Zang 2009), flexural due to exposure of concrete to long periods Contact details
strength (Hsie et al 2008; Lie & Kodur 1996) of heat and sun radiation could have a sig- Unit for Construction Materials, Department of Civil Engineering
Stellenbosch University, Stellenbosch 7602
and reduction in plastic shrinkage cracking nificantly adverse effect on their properties. T: +27 21 808 4445, E: bboshoff@sun.ac.za
(Boshoff & Combrinck 2013) have been The application of macro synthetic fibre is
Keywords: synthetic macro fibres, pull-out load, fibre-reinforced concrete,
reported. The two most widely used fibres typically limited to shotcrete and concrete FRC, single-fibre pull-out, embedment length,
in civil engineering works are steel and floor-on-grade applications, where the critical fibre length, scanning electron microscope, SEM

Odendaal CM, Babafemi AJ, Combrinck R, De Villiers WI, Boshoff WP. Evaluation of locally available synthetic macro fibres in a single-fibre
pull-out test in concrete. J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1744, 10 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a3 21
 F

Frictional
snubbing force
Normal
force

Rocstay Geotex 500 series Geotex 600 series Chryso structural

(30 mm) (50 mm) (50 mm) (50 mm)

Figure 1 Fibre snubbing angle Figure 2 Synthetic (polypropylene) macro fibres

Table 1 Properties of synthetic macro fibres

Fibre Commercial Cross‑sectional Longitudinal lf de(s) de σf
λ
classes name shape geometry (mm) (mm) (mm) (MPa)

Fibre 1 Rocstay X Crimped 30 0.8 0.998 30.1 300

Fibre 2 Rocstay X Crimped 30 0.8 0.701 42.8 300

Fibre 3 Geotex 500 Series Oval Crimped 50 0.9 0.76 65.8 295

Fibre 4 Goetex 600 Series Rectangular Flat 50 0.8 0.907 55.1 275

Fibre 5 Chryso Rectangular Flat 50 0.79 0.615 81.0 336

yield-line design approach (plastic design)
has to be followed to harness the advan-
tages of the fibres (Concrete Society 2014).
There is considerable global scepticism
regarding the use of synthetic fibres for
reinforcing concrete, with South Africa
not being exempted. The scepticism is
strengthened by the fact that synthetic
macro fibres have low stiffness compared
to steel fibres, and have been reported
to creep significantly more than steel
(Babafemi & Boshoff 2015; Pujadas et al
2017). However, this should not discourage
its use, but requires that certain design
guidelines be established for its sensible
application (Pujadas et al 2017).
In this paper, the pull-out performance
of four locally available synthetic macro
fibres in South Africa is evaluated at the
single-fibre level. The main objective of
this study is to assess the fibre-matrix
bond characteristics of these locally
available fibres. The fibre-matrix bond
characteristics would give an indication of
how best each fibre performs. This is done
by performing single-fibre pull-out tests
at different embedment lengths, as well as
different embedment angles to determine
the snubbing effect. The fibre snubbing
effect is the increase in maximum pull-out
force (F) when a fibre is pulled out at an
angle (φ) as opposed to being pulled out in
an aligned fashion, as shown in Figure 1.
The uniqueness of this study is to par-
ticularly investigate these locally available
macro synthetic fibres at the single-fibre
level. Further studies are still ongoing to
relate their response to the macro-level
performance under flexure and shear.
Fibres are randomly distributed in a
fibre-reinforced composite. The fibres in
the composite take different orientation
angles and are embedded at different
lengths across a cracked section. The
overall performance of the composite is
dependent on these factors, and many
more. Therefore the aim of the study (of
the effect of the insertion length and snub-
bing angles at the single-fibre level) is to
assess the individual factors and their con-
tribution to the overall performance of the
reinforced composite. This has aided the
development of models accommodating the
contributions of fibre embedment length
Table 2 Concrete mix design for experimental investigation
Mix
Water Cement
(kg/m3)
Mix 1 240 400
Mix 2 240 480
Mix 3 240 600
and orientation angle in the prediction of
the mechanical response of FRC elements.
However, it is not the goal of this study to
develop any model accommodating these
factors, but to demonstrate that these fac-
tors could have an impact on the overall
performance of FRC elements.
MATERIALS AND CONCRETE MIX
Synthetic macro fibres
Four types of synthetic polypropylene
macro fibres, locally available in South
Africa, were used, as shown in Figure 2.
The fibre properties, both given by the sup-
plier and measured, are shown in Table 1.
Table 1 shows five fibre categories, as
Fibres 1 and 2 were supposed to be the
same, but a significant change in the meas-
ured equivalent diameters was found from
the first batch to the second. Therefore
their specified equivalent diameters are the
Coarse
w/c Sand
aggregate
0.6 782 926
0.5 782 859
0.4 782 759

22 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
same. It is clearly indicated where Fibre 1 Table 3 Fibre embedment lengths tested
or Fibre 2 was used. Embedment length (mm) Fibres 1 & 2 Fibre 3 Fibre 4 Fibre 5
The equivalent diameter (de) of each
L1 10 12.5 12.5 12.5
fibre type was calculated following the
procedure outlined in EN 14889-2:2006 L2 20 25.0 25.0 25.0
(EN 2006) from the measured total fibre L3 30 37.5 37.5 37.5
mass (up to 50 numbers) of each fibre type.
L4 40 50.0 50.0 45.0
Since the fibres do not have a circular
cross-section, the de is the diameter of a
circle having an area equal to the average
cross-sectional area of a fibre, calculated as

400mf 40 mm 20 mm 40 mm
de = (1)
RDπlf (t)

where mf is the total mass of fibres in

grams, lf(t) is the total length of the weighed
fibres in mm (which is determined by mul-
tiplying the fibre length with the number of
fibres) and RD is the relative density of the
fibre (0.91 for polypropylene). In Table 1, lf
is the fibre length, de(s) is the supplied de,
λ the aspect ratio (lf/de) and σf is the sup-
plied tensile strength. It should be noted
that for tests conducted at the single-fibre
level, longer fibre lengths obtained from
the suppliers have been used to be able to Plan view
100 mm
study the effect of the various embedment
Side view
lengths on the pull-out response.

Concrete mix design

The concrete mix constituents are CEM I
52.5N cement with a relative density (RD)
of 3.14, fine natural sand (known locally as
Malmesbury sand) having an RD of 2.62,
and a crushed Greywacke coarse aggregate
(nominal size of 6 mm) with an RD of 2.80.
100 mm

Three concrete mixes (Mix 1, Mix 2, Mix 3),

shown in Table 2, with water-cement ratios
(w/c) of 0.6, 0.5 and 0.4 were used. The
slump was measured in accordance with
SANS 5862-1:2006 (SANS 2006a), and the
compressive strength in accordance with
SANS 5863‑3:2006 (SANS 2006b) at an age
of 28 days.
Concrete specimens were cast without
fibres in the single-fibre pull-out test, except
for the embedded single fibres that were
added after the moulds had been filled.
However, the effect of synthetic macro fibres
on the compressive strength of concrete
was investigated using Fibre 1 (at 0.5, 0.6,
0.7% vol), Fibre 3 (at 0.5, 0.6, 0.7% vol), Fibre
4 (at 0.3, 0.5, 0.6, 0.7, 0.9% vol) and Fibre 5
(at 0.6% vol). It should be noted that Fibre 1
is representative of Fibre 2, hence Fibre 2
was not tested for compressive strength.
A close look at the fibre types (Table 1
and Figure 2) shows that the fibres can be
grouped into three surface configurations:
Front view
Figure 3 Schematics of single-fibre sample preparation (not to scale)
X cross-section and crimped (Fibres 1
and 2), oval cross-section and crimped
(Fibre 3), and rectangular flat (Fibres 4 and
5). The fibre groupings were studied at 0.5%,
0.6% and 0.7% volume, while two additional
boundaries of low and high volumes (0.3% &
0.9%) were tested for the flat fibres.
EXPERIMENTAL METHODS
Single-fibre pull-out tests (SFPOT) were
undertaken to study the pull-out behaviour
Fibre
Concrete
Wooden separator
of the fibres. Fibres 1 and 2 were tested
at an embedment length of multiples of a
third of its length, as well as a 40 mm fibre
length. Fibres 1 and 2 were supplied in two
different diameters and were thus used to
investigate the effect of the diameter on
the pull-out response. The effect of the
variation in fibre diameter was only carried
out for Mix 2 with w/c of 0.5. SFPOT for
Fibres 3, 4 and 5 were performed at embed-
ment lengths of multiples of a quarter
of their lengths, as presented in Table 3,

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 23

Strong
axis
bending
Weak axis bending
Figure 4 C
 ross-section of a flat rectangular
fibre showing strong and weak axes
except for Fibre 5 where the final embed-
ment length tested (L4) was 45 mm.
Furthermore, the snubbing angle effect
of these flexible fibres on the pull-out
load was also investigated. Here only one
embedment length was used for all fibres
(two-thirds of the fibre length for Fibre 1,
which is 20 mm, and half the fibre length
for the others, i.e. 25 mm). Snubbing angles
investigated were 0° (i.e. perpendicular to
the crack plane), 30° and 60°. Only Mix 2
with a w/c of 0.5 was used to investigate
the snubbing effect.
Test specimen preparation
For each fibre type, eight specimens, each
measuring 100 × 100 × 40 mm3 were pre-
pared and tested in the SFPOT. These spec-
imens were obtained by dividing 100 mm
polyvinyl chloride (PVC) cube moulds into
Figure 5 Single-fibre pull-out test setup
24
Table 4 Average compressive strength and slump values
Compressive
Mix w/c
Mix 1 0.4
Mix 2 0.5
Mix 3 0.6
two halves with wooden blocks, resulting in
two specimens per cube mould, as shown in
Figure 3 on page 23. The dry mix constitu-
ents without fibres were mixed for about
three minutes before the addition of water.
After mixing, the fresh concrete was cast
in the moulds and vibrated on a vibrating
table. The fibres were then inserted into
the centre of the fresh concrete up to the
pre-marked embedment lengths. For the
less flexural stiff fibres (Fibres 4 and 5),
one of the flexural stiff fibres (Fibre 1) was
initially used to create the openings before
the less flexural stiff fibres were inserted to
the desired embedment length. The reason
for the difference in flexural stiffness is the
different cross sections of the fibres. After
the insertion, moulds were gently vibrated
to close the voids created between the fibre
and the concrete. The samples were allowed
to set in a temperature-controlled chamber
at 23°C for 24 h, and then submerged in a
curing tank at 22°C until testing at 28 days.
For the evaluation of the snubbing
effect, the fibres were inserted at the snub-
bing angles previously mentioned. The flat
fibres were inserted at a specific orientation
to ensure bending about their strong axis
during pull-out, as shown in Figure 4.
Load cell
LVDT
Drill chuck
Specimen
Hydraulic clamp
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
CoV Slump
strength
(%) (mm)
(MPa)
59.0 4 70
52.2 3 150
43.1 7 200
Test setup for SFPOT
The SFPOT setup is shown in Figure 5.
The tests were performed in a Zwick Z250
Universal Testing Machine. Hydraulic
clamps held the bottom part of the con-
crete specimen. The fibre clamp gripped
the fibre protruding from the concrete, as
close as possible to the concrete surface
to minimise the effect of fibre elongation.
Two 50 mm HBM linear variable differ-
ential transformers (LVDTs) were used to
measure pull-out displacement. The HBM
load cell used has a capacity of 250 kg. The
pull-out tests were performed at a constant
crosshead displacement rate of 0.2 mm/s.
A drill chuck shown in Figure 5 was
used as a fibre clamp for the non-flat
Fibres 1 and 2. The drill chuck was unable
to grip the flat Fibres 4 and 5 without slip-
page, hence a separate clamp was manufac-
tured for gripping these fibres.
This clamp consisted of two flat metal
parts clamping the fibre as shown in
Figure 6. A permanent marker was used
to mark the fibres at the clamp-fibre
interface to help identify fibre slippage.
If fibre slippage was observed at the
clamped section, the results of such tests
were discarded.
Figure 6 C
 lamp developed and used for flat
fibres
RESULTS AND DISCUSSION
45
Compressive strength 0% fibre (control)
40
The results of the compressive strength
and slump tests of mixes without fibres are 35
presented in Table 4, with the coefficient

Compressive strength (MPa)

of variation (CoV) of the compressive 30
strength shown.
As expected, the compressive strength 25
of the mixes increases as the w/c decreases,
20
whereas the slump increases as the w/c
increases. The effect of adding fibres on the 15
compressive strength of Mix 3 is shown for
the four different fibres in Figure 7. Only 10
Fibre 4 was tested at 0.3 and 0.9% dosages,
while Fibre 5 was only tested at 0.6% dos- 5
age. Figure 7 reveals that the addition of
0
the fibres has a negative effect on the com- 0.3 0.5 0.6 0.7 0.9
pressive strength of the concrete at the dos- Fibre dosage (%)
ages investigated. This is consistent with Fibre 1 Fibre 3 Fibre 4 Fibre 5
several works on the effect of synthetic
fibres on the compressive strength of ordi- Figure 7 Influence of fibre type and dosage on compressive strength of concrete
nary concrete (Hsie et al 2008; Babafemi &
Boshoff 2015).
The crimped fibre (Fibre 3) shows an 140
increase in compressive strength from the
120
0.5 to 0.7% dosages, even attaining the
strength of the control at 0.7% fibre dosage. 100
Pull-out force (N)

For the slightly crimped (Fibre 1) and the

flat (Fibre 4) no particular trend is apparent. 80
At 0.6% fibre dosage, all fibres resulted in
60
a lower compressive strength compared to
the control, with Fibre 1 showing the most 40
severe reduction. The reduction in compres-
sive strength of concrete due to the addition 20
of fibres is attributed to the increased
0
voids created during mixing and the non- 0 5 10 15 20 25 30
homogeneity in the microstructure due to Fibre pull-out (mm)
the distribution of fibres within the matrix. Fibre 1 Fibre 3 Fibre 4 Fibre 5

Pull-out behaviour of synthetic Figure 8 Typical pull-out responses of each fibre type
macro fibres
All results presented here are the average
of the maximum pull-out load for each set 250
of fibre type and embedment length. This
Average maximum pull-out force (N)

value is typically at the onset of pull-out 200

and gradually drops until complete pull-out
is achieved. A typical pull-out response for
150
each fibre type is shown in Figure 8.

Fibres 1 and 2 100

The average maximum pull-out loads of
Fibre 1 at different embedment lengths and 50
w/c are shown in Figure 9. Also shown is
the response of Fibre 2, which is the same
0
as Fibre 1 except for a smaller equivalent 0 10 20 30 40
diameter at different embedment lengths, Embedment length (mm)
and a w/c of 0.5. Fibre 1, w/c = 0.4 Fibre 1, w/c = 0.5 Fibre 1, w/c = 0.6 Fibre 2, w/c = 0.5
One significant factor that influences
the average maximum pull-out loads, as Figure 9 Average pull-out loads of Fibres 1 and 2 at different embedment lengths and w/c

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 25

250
Average maximum pull-out force (N)
200
150
100
50
0
0 10
Embedment length (mm)
Fibre 3, w/c = 0.4
Figure 10 Average pull-out loads of Fibre 3 at different embedment lengths and w/c
250
Average maximum pull-out force (N)
200
150
100
50
0
0 10
Embedment length (mm)
Fibre 4, w/c = 0.4
Figure 11 Average pull-out loads of Fibre 4 at different embedment lengths and w/c
can be seen in Figure 9, is the fibre diam-
eter. With a 40 mm embedment length, a
significant drop in the average maximum
pull-out load is observed. It should be men-
tioned that the fibres with 40 mm embed-
ment lengths had an equivalent diameter
similar to Fibre 2 which is less than Fibre 1,
which explains the drop in the average
maximum pull-out load at this embedment
length. Similarly, Fibre 2 with the lesser
equivalent diameter showed a significantly
lower maximum pull-out load compared
to Fibre 1 at embedment lengths of 20 mm
and 30 mm. This indicates that the average
maximum pull-out load is a function of the
equivalent diameter of the fibre. However,
the effect of the equivalent diameter for
the flat fibres still needs to be researched.
The w/c showed no effect on the pull-out
response. Figure 9 did show some effect of
the w/c, but this is not significant as the
highest and lowest w/c showed almost the
26
20 30 40 50
Fibre 3, w/c = 0.5 Fibre 3, w/c = 0.6
20 30 40 50
Fibre 4, w/c = 0.5 Fibre 4, w/c = 0.6
same behaviour, which is not expected if
the w/c did have a significant effect.
The results indicate that with the
increase in the fibre embedment length, the
higher the average maximum pull-out load
will be at various w/c. The increase in the
average maximum pull-out load as the fibre
length increases for this type of fibre has
also been reported elsewhere (Babafemi &
Boshoff 2017).
Fibre 3
The effect of embedment length and w/c
on the average maximum pull-out load of
Fibre 3 is shown in Figure 10. Fibre 3 shows
that the pull-out performance is not a
function of the w/c, as all performed simi-
larly at different w/c. In addition, the aver-
age maximum pull-out load also increases
with increasing embedment length.
In Figure 10, beyond an embedment
length of 37.5 mm, a drop in the average
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
maximum pull-out load can be observed.
It should be mentioned that, from a physi-
cal observation of the 50 mm fibres, they
showed a less pronounced crimped config-
uration (lower amplitude) compared to the
shorter fibres. It is believed that the drop
in the average maximum pull-out load at
this embedment length is due to the lower
amplitude of the fibre. Bentur et al (1997)
and Won et al (2006) also showed that the
amplitude of a polymeric fibre has an effect
on the average maximum pull-out load. All
lengths of Fibre 3 have the same equivalent
diameter, hence the effect of equivalent
fibre diameter plays no role in the observed
drop at 50 mm as explained for Fibre 1.
Fibre 4
The pull-out performance of Fibre 4, which
has a flat longitudinal geometry, is also not
influenced by the w/c, while the average
maximum pull-out load also increases with
an increase in embedment length, as shown
in Figure 11. The average maximum pull-
out load is much lower than for Fibres 1, 2
and 3. This suggests that the fibre geometry
is an important factor in the pull-out
performance of synthetic macro fibre when
used in concrete. However, increase in the
average maximum pull-out load can be
observed up to a fibre embedment length of
50 mm. The near linear average maximum
pull-out load to embedment length relation-
ship observed for Fibre 4 indicates that the
critical fibre length was not embedded in
the matrix as only a few specimens experi-
enced fracture. The critical fibre length is
the length for which a shorter length would
result in no fibre rupture and a larger length
would result in significant fibre rupture
during mechanical testing. The critical fibre
length is double the critical embedment
length, as determined using single-fibre
pull-out tests. Using the critical length for
fibres will result in the optimal behaviour.
Fibre 5
The average maximum pull-out load of
Fibre 5, as with all other synthetic macro
fibres, is dependent on the embedment
length, with no visible effect of w/c, as
shown in Figure 12. It can be concluded
that the average maximum pull-out load
of synthetic macro fibres is not dependent
on w/c ranging from 0.4 to 0.6. The average
maximum pull-out load for Fibre 5 is 111 N,
which is similar to the pull-out performance
of Fibre 4 that also has a flat geometry.
A near linear increase in average
maximum pull-out load with increasing
embedment length occurred up to around
37.5 mm, where fibres fractured. At embed- 250
ment lengths of 45 mm, all but two of the

Average maximum pull-out force (N)

fibres fractured across all w/c investigated. 200
The average maximum pull-out loads
appears to plateau at around 110 N.
150
Figure 13 shows the percentage of fibres
that fractured for all fibres tested at different
embedment length. During the pull-out 100
test of Fibre 1, complete fibre pull-out was
recorded up to an embedment length of 50
30 mm. It should be noted that Fibre 1 (with
the greater de) was only tested up to 30 mm
0
embedment length as previously described, 0 10 20 30 40 50
as the 40 mm embedment length had a Embedment length (mm)
smaller de. However, the 40 mm embedded Fibre 5, w/c = 0.4 Fibre 5, w/c = 0.5 Fibre 5, w/c = 0.6
fibres of Fibre 1 with a smaller de showed
a fracture of 50% (Figure 12). Meanwhile, Figure 12 Average pull-out loads of Fibre 5 at various embedment lengths and w/c
complete pull-out was only recorded at an
embedment length of 10 mm for Fibre 2
(with the lesser de). At 20 mm embedment 100
length for Fibre 2, fracture was about 17%,
while at 40 mm embedment lengths, about 80
54% fractured during pull-out. The result
Fibre fracture (%)

of the percentage fracture at an embedment

60
length of 40 mm for the fibres with lesser
de grouped with Fibre 1, agrees much better
with similar fibres classified as Fibre 2 at 40
the same embedment length. These results
indicate that the critical fibre length (length 20
at which fracture occurs) is a function of the
equivalent fibre diameter for the crimped
0
fibres tested. Thus, if a significant portion 0 10 20 30 40 50
of fibre fractures at a particular embedment Embedment length (mm)
length, i.e more than 10%, it indicates that Fibre 1 Fibre 2 Fibre 3 Fibre 4 Fibre 5
the critical embedment length was reached
between the previous embedment length Figure 13 Percentage of fibre fractures at different embedment lengths
and the current embedment length.
Less than 40% fracture occurred for
Fibre 3 at 37.5 and 50 mm, while less 2.0
Normalised maximum pull-out load (N)

than 10% occurred at 37.5 and 50 mm for

Fibre 4, and about 90% at 37.5 and 45 mm 1.6
for Fibre 5. It can be inferred that the
embedment lengths at which the peak per-
1.2
centage of fractures occur is an indication
of the critical fibre length. Fibre pull-out
is the preferred failure mechanism rather 0.8
than fibre rupture, as pull-out results in
a ductile failure, with the fibre pull-out 0.4
process absorbing more energy than a
fibre rupture. The critical fibre length is
0
twice the critical embedment length, as the 0 10 20 30 40 50 60
shortest embedment length is dominant Snubbing angle (°)
when a fibre crosses a crack plane in Fibre 1 Fibre 3 Fibre 4 Fibre 5
concrete. It is important to note that the
fibre fracture did not occur at the fibre’s Figure 14 Snubbing effect on pull-out load of synthetic macro fibres
clamping point. For samples where fracture
occurred at the clamp, the results were dis- not exceed the fibre’s stress. On the other It is common knowledge that the w/c
carded. Evidently, the fibres that pulled out hand, the fracture fibres show that the fibre influences the compressive strength of
completely show that the pull-out stress did stress is exceeded during pull-out. concrete. The lower the w/c, the higher

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 27
the compressive strength. This, how-
Description Image(s)
ever, does not influence the fibre/matrix
interface, and is believed to be due to the
hydrophobic nature of the polypropylene
fibres. The poor bonding is enhanced by
the fibre geometry rather than the concrete Undamaged
strength, which is dependent on the w/c.
In fact, the maximum fibre/matrix bond
strength has been reported to be attained
just within the first two days (Singh et al
2004), unlike the compressive strength that
continues to increase with age. Therefore
the w/c does not influence the maximum Pull-out damage
bond load, rather, the maximum bond load
is a function of the properties of the fibre.

Effect of snubbing angle

on pull-out load Clean Tearing
The orientation of fibres influence the pull- breakage
out performance of cement composites.
Fracture damage
Figure 14 shows the normalised plot of the
average maximum pull-out load against
the snubbing angles of all fibres tested.
All the fibres had an embedment length of
25 mm, except Fibre 1 which was embed- Figure 15 SEM images of Fibre 1 before and after pull-out
ded at 20 mm. The purpose of these tests
is to evaluate the effect of snubbing for
Description Image(s)
different fibres, therefore this small change
in embedment length for the one fibre is
believed to be insignificant.
As the results were normalised to the
Undamaged
pull-out resistance of each fibre type at 0°,
they all start at 1.0. At 30° and 60° differen­
ces can be seen for the various fibres, indi-
cating increased frictional effect at these
angles. Fibre 4 experienced a 39% and 70%
increase in the average maximum pull-out
loads at 30° and 60° respectively. As men- Pull-out damage
(middle and
tioned, the bending of the fibre was about end)
its strong axis, as shown in Figure 3. If the
fibre is considered as a beam subjected to
flexural bending, bending the fibre about
its strong axis requires a larger force than
for bending it to the same rotation about
the weak axis. This could subsequently lead
Pull-out damage
to a larger friction over a constant area for (interface)
strong axis bending as opposed to weak
axis bending. In addition, strong axis bend-
ing is exerted over a smaller area, which
is believed will lead to a larger frictional
stress. As the increase in the snubbing
angle causes an increase in the frictional
force over a constant surface area, a larger
increase in the overall pull-out load contri-
Fracture damage
bution is believed to occur.
Fibre 5 experienced a 7% and 10% pull-
out load increase at 30° and 60°, respectively.
This plateau can be attributed to the fibres
rupturing when the tensile strength of the
fibres is exceeded due to the additional Figure 16 SEM images of Fibre 3 before and after pull-out

28 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering

Description
Undamaged
Pull-out damage
(middle and
end)
Figure 17 SEM images of Fibre 4 before and after pull-out
Description
Undamaged
Pull-out damage
Fracture damage
(in concrete and
in clamp)
Figure 18 SEM images of Fibre 5 before and after pull-out
resisting force offered by the frictional
snubbing resistance. This is confirmed
by six of the eight fibres fracturing at the
cement-matrix surface at the 60° snubbing
angle. While the crimped Fibre 3 showed a
25% increase at 60° snubbing angle, Fibre 1
showed an increase of 11% at 60°. At 30°,
Fibre 1 showed a decrease in pull-out load
of 13% before increasing again at 60°. This is
likely due to inconsistent orientation of the
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
Image(s)
Image(s)
cross-sectional axis which the fibre was bent
about. Overall, the flat fibres performed bet-
ter than the crimped fibres at the snubbing
angles considered.
SEM photos of pulled out fibres
Scanning Electron Microscope (SEM)
images were taken for each type of fibre to
study the effect of friction during the pull-
out testing. Fibre 1 showed pronounced
surface damage that can be attributed to its
X-shaped cross section, hence enhancing
bonding between the fibre and matrix.
Fibre fracture of Fibre 1 appears to be a
combination of clean breakage and tearing,
as shown in Figure 15.
Fibre 3 shows severe fibre damage
caused by the scraping of the cement
matrix against the fibre during pull-out,
as shown in Figure 16. Variable damage
can be observed at different portions of
the fibre. Significantly less damage at the
cement matrix interface or surface can be
observed as opposed to the images at the
middle of the fibre and at the embedded
ends. The increased damage is due to the
embedded end having to travel further to
exit the cement matrix, resulting in more
scrapings. The fracture mode for Fibre 3
appears to be fibre splitting.
Fibre 4, which showed the lowest aver-
age maximum pull-out load, also showed
minimal surface damage, as shown in
Figure 17. This is due to the flat geometry
of the fibre.
Severe scrapings after pull-out were also
observed for Fibre 5 as shown in Figure 18.
However, the damage is more than
observed for Fibre 4, indicating better bond
than Fibre 4. This is also reflected by the
average maximum pull-out load obtained
in the single-fibre pull-out tests. Fibre 5
also tends to fibrillate after pull-out.
CONCLUSIONS
The pull-out performance of four locally
available synthetic macro fibres was deter-
mined using single-fibre pull-out tests. The
effect of these fibres on the compressive
strength was also tested. The following
conclusions can be drawn from this study:
■■ Synthetic macro fibres do not contribute
to the compressive strength of concrete.
■■ The pull-out performance of synthetic
macro fibres in a single pull-out test is
not a function of w/c.
■■ The fibre diameter does influence the
average maximum pull-out load. Fibres
with greater equivalent diameter show
higher pull-out loads for the fibres test-
ed. More work is required to determine
similar trends for other types of fibres.
■■ The average maximum pull-out load of
synthetic macro fibres increases as the
embedment length increases, while fibre
geometry has a significant influence.
■■ The average maximum pull-out load of
synthetic macro fibres is positively influ-
enced with an increase in the snubbing
29
 angle, where in addition flat fibres per-
form better than crimped fibres.
■■ Fibre 1 performed optimally in terms
of the average maximum pull-out load
attained. This is due to its crimped sur-
face configuration and X-shaped cross
section, which enhanced mechanical
interlocking with the matrix.
■■ The Scanning Electron Microscope
images of fibres taken after pull-out
show that fibre geometry has an effect
on the maximum pull-out load. The
more irregular the fibre geometry, the
more the frictional effect during pull-
out leading to higher pull-out load.
REFERENCES
Alani, A M, Beckett, D 2013. Mechanical properties
of a large-scale synthetic fibre reinforced concrete
ground slab. Construction and Building Materials,
41: 335–344.
Babafemi, A J & Boshoff, W P 2015. Tensile creep of
macro-synthetic fibre reinforced concrete (MSFRC)
under uni-axial tensile loading. Cement and
Concrete Composites, 55: 62–69.
Babafemi, A J & Boshoff, W P 2017. Pull-out response
of macro synthetic fibre from concrete matrix:
Effect of loading rate and embedment length.
Construction and Building Materials, 135: 590–599.
Bentur, A, Peled, A & Yankelevsky, D 1997. Enhanced
bonding of low modulus polymer fibers-cement
matrix by means of crimped geometry. Cement and
Concrete Research, 27(7): 1099–1111.
Bernard, E S 2004. Creep of cracked fibre reinforced
shotcrete panels. Proceedings, 2nd International
30
Conference on Engineering Developments
in Shotcrete (Shotcrete: More Engineering
Developments), 47–57.
Boshoff, W P & Combrinck, R 2013. Modelling the
severity of plastic shrinkage cracking in concrete.
Cement and Concrete Research, 48: 34–39.
Buratti, N, Mazzotti, C & Savoia, M 2011. Post-
cracking behaviour of steel and macro-synthetic
fibre-reinforced concretes. Construction and
Building Materials, 25(5): 2713–2722.
Concrete Society 2014. Concrete Industrial Ground
Floors. A Guide to Design and Construction, 4th ed.
Technical Report 34, Camberley, UK: The Concrete
Society.
De la Fuente, A, Escariz, R C, de Figueiredo, A D
& Aguado, A 2013. Design of macro-synthetic
fibre reinforced concrete pipes. Construction and
Building Materials, 43: 523–532.
EN (European Standard) 2006. BS EN 14889-2:2006.
Fibres for concrete. Polymer fibres. Definitions,
specifications and conformity. London: British
Standards Institution.
Ferrara, L & Meda, A 2006. Relationships between
fibre distribution, workability and the mechanical
properties of SFRC applied to precast roof elements.
Materials and Structures, 39(4): 411–420.
Hannant, D J 1998. Durability of polypropylene fibers
in Portland cement-based composites: Eighteen
years of data. Cement and Concrete Research, 28(12):
1809–1817.
Hsie, M, Tu, C & Song. P S 2008. Mechanical
properties of polypropylene hybrid fiber reinforced
concrete. Materials Science and Engineering. A.
Structures, 494(1–2): 153–157.
Lie, T T & Kodur, V K 1996. Thermal and mechanical
properties of steel-fibre-reinforced concrete at
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
elevated temperatures. Canadian Journal of Civil
Engineering, 23(2): 511-517.
Mindess, S & Zhang, L 2009. Impact resistance
of fibre-reinforced concrete. Proceedings of the
Institution of Civil Engineers – Structures and
Buildings, 162(1): 69–76.
Pujadas, P, Blanco, A, Cavalaro, S, de la Fuente A &
Aguado, A 2017. The need to consider flexural post-
cracking creep behavior of macro-synthetic fiber
reinforced concrete. Construction and Building
Materials, 149(15): 790–800.
Richardson, A E 2005. Bond characteristics of
structural polypropylene fibres in concrete with
regard to post-crack strength and durable design.
Structural Survey, 23(3): 210–230.
Singh, S, Shukla, A & Brown, R 2004. Pullout
behaviour of polypropylene fibers from cementitious
matrix. Cement and Concrete Research, 34: 1919–25.
Sorelli, L G, Meda, A & Plizzari, G A 2006. Steel fiber
concrete slabs on ground: A structural matter. ACI
Structural Journal, 103(4): 551–558.
SANS (South African National Standard) 2006a. SANS
5862-1:2006. Concrete Tests – Consistence of Freshly
Mixed Concrete – Slump Test. Pretoria: South
African Bureau of Standards.
SANS (South African National Standard) 2006b. SANS
5863-3: 2006. Concrete Tests –Compressive Strength
of Hardened Concrete. Pretoria: South African
Bureau of Standards.
Wang, Y, Backer, S & Li, V C 1987. An experimental study
of synthetic fibre reinforced cementitious composites.
Journal of Materials Science, 22(12): 4281–4291.
Won, J P, Lim, D H & Park, C G 2006. Bond behaviour
and flexural performance of structural synthetic
fibre-reinforced concrete. Magazine of Concrete
Research, 58(6): 401–410.
The effect of aeration TECHNICAL PAPER
through an internal gallery Journal of the South African
Institution of Civil Engineering

of a dam on the cavitation

ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 31–43, Paper 1727

risk of Roberts splitters GERARD CALITZ obtained his MEng in Civil

Engineering cum laude from Stellenbosch
University in 2016. He currently works in
Aurecon’s Pretoria office as a graduate civil
G Calitz, G R Basson engineer specialising in the design and
detailing of hydraulic structures. His
research interests include dams, dam
spillways and open channel hydraulics.
Roberts splitters is an effective means of energy dissipation for dam spillways. Roberts’ (1943)
standard unaerated splitter design procedure is, however, limited to a spillway head (H) of 3.0 m Contact details:
Aurecon Centre
(q ≈ 12 m2/s). In order to avoid cavitation at higher design spillway heads, this study investigated Lynnwood Bridge Office Park
the artificial aeration of the flow by local air vents positioned on the splitters. A 1:20 scale 4 Daventry Street
hydraulic model of an ogee spillway equipped with Roberts splitters was constructed. Two Lynnwood Manor
Pretoria 0081
aerated models, with differently sized air vents, were compared to an unaerated control model
South Africa
in order to determine the effect that the proposed aeration system has on the cavitation risk of T: +27 12 427 2823
the splitters at prototype unit discharges (q) of up to 50 m2/s. E: gerard.calitz@aurecongroup.com
At the maximum tested spillway head of 7.6 m (q = 50 m2/s) the minimum pressures and air
concentration around the splitters of both aerated models increased considerably. It was further PROF GERRIT BASSON Pr Eng (FSAICE) is
Head of the Water Division in the
observed that the unaerated splitters were prone to drowning at high spillway heads, leading to
Department of Civil Engineering at
unfavourable hydraulic conditions. Stellenbosch University, South Africa. He
Based on the results of this study, the addition of aeration through an internal aeration gallery specialises in river hydraulics, sedimentation
can increase the unit discharge capacity of Roberts splitters to at least 50 m2/s, up by 43%, from and design of hydraulic structures. He has
30 years’ experience in the design of
the unaerated limit of 35 m2/s.
hydraulic structures and is well known
internationally as the recent past vice-president of ICOLD (International
Commission on Large Dams) and of WASER (World Association of
Introduction are typically used on high dams where the Sediment and Erosion Research). He has work experience from 21
countries and has been involved in several missions of the World Bank as
spillway flow velocities are too fast for a expert.
Roberts splitters as energy stilling basin, or the unit discharge too high
Contact details:
dissipator for a stepped spillway. The Roberts splitters Department of Civil Engineering
Roberts splitters are a type of energy-dissi- system consists of a series of projecting Stellenbosch University
pating measure located near the top of dam teeth or splitters immediately upstream of Private Bag X1
Matieland 7602
spillways, as can be seen on the Vanderkloof a continuous lip or step (Mason 1983). If
South Africa
Dam spillway in Figure 1, originally devel- designed correctly, two sets of jets created T: +27 21 808 4355
oped by Lt Col DF Roberts for the Loskop by the splitters and step will collide in mid- E: grbasson@sun.ac.za
Dam in 1936 (Roberts 1943). These splitters air, breaking up into spray, and dissipating

Keywords: Roberts splitters, dam spillway, energy dissipation,

Figure 1 R
 oberts splitters on the Vanderkloof Dam spillway (vanderkloofdam.com 2014) artificial aeration, cavitation

Calitz G, Basson GR. The effect of aeration through an internal gallery of a dam on the cavitation risk of Roberts splitters.
J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1727, 13 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a4 31
the energy through air resistance (Roberts
1980), as can be seen in Figure 2.
During operation, the intended goal of
the splitters is to split the sheet of water
flowing down the spillway. By doing this,
the direction of the flow is rapidly changed
and flow separation may exist around the
splitters, leading to sub-atmospheric pres-
sures. If the pressure in the fluid becomes
too low (approaches vapour pressure),
cavitation may occur, which could cause
serious damage to the spillway structure.
Cavitation in general can be mitigated
in two ways: firstly, by ensuring that the
fluid pressure remains above 3 m absolute
(Chadwick et al 2013), and secondly, by
introducing at least 8% air concentration
into the flow (Chanson 1992). On a dam
spillway with Roberts splitters, air can be
introduced to the water flow at atmos-
pheric pressure via air vents connected to Figure 2 R
 oberts splitters in operation on the Gariep Dam spillway (Calitz & Basson 2015)
an atmospheric air source. This is called
artificial aeration. This air flow to the
air vents needs not be pumped or pres-
Bridge pier
surised, if properly designed, because sub-
atmospheric pressures within the nappe Splitter aeration gallery
created by the separated flow over splitters Gallery Lateral air vent 00.3

9.0
will naturally suck air out of the vents. This 3.2 3.4
is true provided that the pressure of the air Air vent 00.6
3.4

3.2
in the cavity of the aforementioned nappe 8.54
Step
is lower than the air in the air vent (which
should be atmospheric or higher). Although Air vent 00.9 6.71
aeration has previously been provided for 00.9
the Roberts splitters, for example on the
Joint

Joint
spillways of the Gariep and Vanderkloof
Dams (both on the Orange River in South
Africa), “the need or effectiveness of aera- Section through crest Front elevation
tion by internal ducts has, however, not
been proved conclusively” (Jordaan 1989). Figure 3 T ypical detail of Gariep Dam splitter aeration (Roberts 1977)

Objective of the study
The main objective of the study was to
determine what effect the artificial aeration
of Roberts splitters (through an internal
gallery) has on the local negative pressures
around the splitters, with the goal of alle-
viating cavitation risks at prototype unit
discharges of up to 50 m2/s.
To achieve the objective, physical
hydraulic model tests were conducted in the
hydraulic laboratory of the Civil Engineering
Department of Stellenbosch University.
Limitations of the study
The maximum discharge that was evalu-
ated was limited by the model scale and
the laboratory’s pump capacity. The scale
had to be large enough to construct each
splitter with the necessary detail to apply
pressure-monitoring instruments.
The study served mainly as an investi-
gation into the improvement of the original
Roberts procedure for the design of crest
splitters to dissipate the energy of a flood
(Roberts 1943), with the goal to alleviate
cavitation risks. Roberts’ splitters did not
have any form of air vents or ducts, and
thus, with the addition of such features, a
different optimisation might be needed.
The downstream effect of the aeration was
not evaluated, including energy dissipation
and apron pressures. As mentioned by
Mason (1983), until a comprehensive study
of these crest splitters is done to point
towards a set of design guidelines, further
optimisation of the splitter configuration
would be needed, especially at the high
spillway heads we see on modern dams.
In lieu of Roberts’ spillway head limita-
tion of 3 m (q = 12 m2/s), again, optimisation
of the crest and step configuration might
be needed for a design head of 6.7 m
(q = 40 m2/s, as used during the study). It
could therefore not be assumed that the dis-
persive action of the splitters would stay the
same as for those with a lesser design head.
During the study, the size of the air vents
was limited by the space available between
the pressure sensors on the downstream
end of the splitter. Thus, only two sizes were
investigated. This limited a broad recom-
mendation on the optimal air vent size.
Relevant literature
Limitations to Roberts splitters
Roberts (1943) first introduced the limit of
3.0 m of spillway head. He stated that indi-
vidual model tests were required to validate

32 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering

Flow
Aeration
gallery – not
open to
atmosphere Air
Air
Safety screen
Air
Figure 4 W
 orking of Gariep Dam splitter
aeration system
his design if the design head (HD) of a given
prototype was greater than 3.0 m. If using
a conventional ogee spillway, this typically
equates to a unit discharge of 12 m2/s.
This rule was followed for all subsequent
dam designs until that of the Gariep and
Vanderkloof Dams.
With the introduction of artificial
aeration and the dam type being a double
curvature arch dam, the distance the jet
had to travel away from the toe of the dam
decreased considerably, and the allowable
spillway heads (considering future raising)
were tripled to 9.1 m and 9.0 m respectively
(Jordaan 1989). The splitters were, however,
dimensioned for a spillway head of 7.3 m and,
while it was discovered that sub-atmospheric
pressures existed on the spillway just down-
stream of the crest at heads greater than
6.0 m, the cavitation coefficient of σ = 0.3 was
considered acceptable (Back et al 1973).
Roberts (1943) further stated that a
critical head (HC) exists at which the split-
ters become drowned. He designated this
0600 mm inlet pipe
Qin
Baffle wall
1.6
0.6 1.2
Figure 5 G
 eneral layout of the hydraulic models (illustration not to scale)
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
as 1.2 times the design head. Further tests
at the Rhenosterkop Dam in Mpumalanga,
South Africa, confirmed Roberts’ initial
suggestions (Jordaan 1989).
Existing aeration of Roberts splitters
The splitters and aeration of the Gariep
and Vanderkloof Dams were similarly
and simultaneously designed and tested
by the Société Grenobloise d’Etudes et
d’Applications Hydrauliques (Sogreah, now
Artelia) in Grenoble, France (PJ Mason
2016, personal communication, 24 July).
It consists of two 0.6 m diameter air vents
at the end of each splitter along with a
single 0.3 m diameter lateral air vent on
both sides of each splitter. A series of
larger 0.9 m diameter vents are located
below the continuous step downstream of
the splitters (see Figure 3). According to
Mason (2016, personal communication, 24
July), the splitter air vents are fed by the
larger intake vents set in the step below,
hence circulating air locally for each split-
ter (Figure 4). The aeration gallery is not
open to the atmosphere; therefore the air
needed to aerate the flow only comes from
the intake vents in the step. The aeration
gallery rather serves to connect the step’s
large air vents with the smaller vents of the
splitter, and for inspection and access to
the splitters. Roberts (1977) mentioned that
the aeration of the Gariep Dam’s splitters
improved the flow stability and reduced
vibration noise.
During previous studies at Stellenbosch
University (Calitz 2014; Langa 2015) it
was found that water would flow into the
Water level needle
Plastered wall
9.1
24.5
Model layout: Longitudinal section
air vents on the splitters at high spillway
heads and cause potential air flow blockage
within the system. Thus, this study was
further concerned with investigating an
aeration system whereby air would solely
be provided via the internal aeration gal-
lery, with a separate option for draining
any excess inflow water.
Hydraulic model
Scope of the hydraulic model
A 1:20 physical hydraulic model of an
ogee spillway with three configurations of
Roberts splitters were constructed:
1. A control model of unaerated Roberts
splitters, intended to measure the pres-
sure and air concentration around a
splitter and justify the ideal positioning
of air vents on the splitters
2. A first aerated model of Roberts split-
ters, containing small air vents, intend-
ed to alleviate cavitation risks of the
splitters and analyse the performance of
the proposed aeration gallery
3. A second aerated model of Roberts
splitters, containing larger air vents,
intended to further improve the flow
conditions as stated above, and compare
aeration performance with that of the
first aerated model.
In each case, the spillway was placed with-
in a 24.5 m long and 1.2 m wide plastered
brick flume. Figure 5 shows the general
layout of the hydraulic models and Figure 6
shows a photograph of the ogee spillway
with aerated Roberts splitters in operation.
Perspex viewing
pane
Qout
Drainage channel
33
 Figure 7 shows a close-up of the centre
splitter of the model. This splitter was used
to measure the pressure and air concentra-
tion of the flow around it (all surfaces below
the horizontal top surface) and analyse
the performance of the proposed aeration Air intake
Splitters
system. Note the definition of faces of the
splitter, as used for the entirety of this paper.

Model scale and laboratory limitations

The scale of the model was selected as
1:20. This was to allow a prototype design
unit discharge of 40 m2/s and a maximum
unit discharge of 50 m2/s within the limits
of the laboratory, whilst avoiding scale
effects due to viscosity and surface tension
(Robertson 2014). The model was scaled in
accordance with the Froude Law, as gravi- Aeration duct
tational and inertial forces were dominant
(Bosman & Basson 2012), and was built as Figure 6 P
 hotograph of the first aerated hydraulic model
small as possible to maximise the model
discharge that could be achieved from the
laboratory’s supply limit of 700 ℓ/s.
Side face
Design and construction

Ogee spillway and Roberts splitters

The hydraulic model consisted of an ogee
End face
spillway, Roberts splitters and an aeration
gallery. As mentioned above, the prototype
design unit discharge (q) was chosen as
40 m2/s, due to laboratory limitations, which
equated to a design head (HD) of 6.79 m.
The ogee spillway was designed accord- Side bottom Bottom face
ing to the USBR’s (United States Bureau face
of Reclamation) Design of Small Dams
(USBR 1987), which allows the design head
(H0) used to determine the ogee’s crest Figure 7 D
 efinition of splitter faces
shape to be 75% of the maximum head
(He) to pass over the spillway. This will the crest shape was calculated from HD The procedure’s main input variable to
produce sub-atmospheric pressures equal as 5.09 m. produce the splitter configuration is the
to about half the design head (p ≈ –H0/2) The model’s splitters were designed relevant design spillway head (HD) of 6.79 m.
on the crest. Accordingly, the prototype according to the original Roberts (1943) The control model’s splitters were unaer-
ogee design head (H0) used to determine procedure, as simplified by Roberts (1980). ated, whilst the first and second aerated
models contained air vents in the splitters.
Table 1 Model design parameters according to USBR (1987) The general design parameters of the ogee
Prototype Model spillway are summarised in Table 1, and the
Output parameter Symbol Unit
value value dimensions of the splitters in Table 2.
Design unit discharge q m2/s 40.0 0.447
Air vents and aeration gallery
Design discharge Q m3/s 960.0 0.537 Air vents were placed on the two side
Effective spillway length Leff m 24.0 1.2 faces and the downstream end face of the
splitters where the lowest pressures were
Spillway discharge coefficient Ce 2.262
measured during the unaerated model tests
Design head HD m 6.79 0.339 (see Figure 8). The existing aeration sys-
Ogee design head used to determine tems of the Gariep and Vanderkloof Dams
H0 m 5.09 0.255
the crest shape (= 0.75 × HD) were taken into consideration and the ratio
Dam height P m 20.8 1.039 of these dams’ design spillway head-to-vent
sizes was used in dimensioning the model’s
Downstream slope (H : V) 0.75 : 1 0.75 : 1
air vents. The design head (H0) used for the

34 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Table 2 S plitter dimensions according to Table 3 Dimensions of splitter air vents
Roberts (1943)
First aerated model Second aerated model
Prototype Model Parameter
Splitter Model (mm) Prototype (m) Model (mm) Prototype (m)
value value
dimension
(m) (mm) Ømain 20.0 0.4 23.0 0.46
Height below
10.964 548.2 Øside 10.0 0.2 11.5 0.23
spillway crest

Splitter width 2.575 128.8

Splitter spacing 3.425 171.2

Side air vents
Splitter length 3.425 171.2

Step length 4.800 240.0

dimensioning of the Gariep Dam’s splitters

was 7.3 m, and the main air vent diameter
(Ømain) is 0.6 m. This gives Ømain/H0 equal
to 0.08. Using this ratio on the ogee design
head of the model (H0) of 5.09 m gives a
prototype main air vent diameter (Ømain)
of 0.41 m. This is equivalent to a model
diameter of 20.5 mm, but was rounded to
20 mm for ease of construction. The side
vents of the Gariep and Vanderkloof Dams
Main air vents
are half the size of the main vents, and
the model’s side vents were accordingly
dimensioned. Figure 8 P
 ositions of the air vents
The second aerated model’s larger air
vents were limited to 23 mm by the space
available between the pressure sensors on 15
64.38 25 25 14.38
the downstream face of the model splitters 30 30 30 30 23.49

(see Figure 9). This is equivalent to a pro-

12 18 18 12
12 18 18 12

A1 A2 A3 E5 E4 E3 E2 E1
totype diameter of 0.46 m, a 15% increase
B1 B2 B3 E6
from that of the first aerated model. If larger
25

C1 C2 C3 E7
air vents were to be tested, several pressure
25

sensors would have had to be removed,
jeopardising the comparability of the results.
The dimensions of the air vents are
summarised in Table 3, where Ømain and
Øside are the diameters of the main and
side vents respectively.
All air vents were fed from one side by a
single aeration gallery. The aeration gallery
was not open on both sides due the layout
of the model’s flume. The model’s aeration
gallery was designed to be large enough to
allow the air vents to act as the control of
the air flow, and to avoid any significant
variability in air flow to each splitter. This
meant that the cross-sectional area of the
duct had to be significantly more than the
accumulated area of all the air vents of the
second aerated model.
The required accumulated area of all
four splitters’ prototype air vents was cal-
culated as 1.662 m2. The constructed aera-
tion duct had a prototype cross-sectional
area of 3.801 m2, which included a safety
factor of 2.0. The aeration duct included a
drainage port and end cap to allow water
D3 F3 F2 F1
D1 D2
D4
F4 30
D5
30
34.98 58.8 34.98
Front view Side view
Figure 9 S plitter pressure sensor positions
entering the system through the air vents symmetrically placed on the central splitter
to be drained when needed. on the spillway, measuring the pressure on
four of its faces. This could be mirrored
Measuring techniques to include all six downstream faces of the
and equipment splitter. The locations of the sensors are
shown in Figure 9. Readings were taken
Pressure readings on splitters at all locations for a duration of three
Fourteen Wika S-10 high-quality pressure minutes at a frequency of 100 Hz. Through
transmitters were used to measure the statistical analysis and recommendations
dynamic fluid pressure at several positions by Calitz (2015), the minimum pressure
on the splitters. These were connected to from each data set was selected as the value
the two measuring splitters by Ø3 mm out- exceeded 99.85% of the time.
side diameter plastic tubes. The transmit-
ters had a range of ±100 mbar, an accuracy Air concentration around splitters
of ±0.2% and a reading repeatability of To analyse the cavitation risk of the splitters,
±0.1%. The pressure sensors’ tubes were the air concentration (Air%) of the flow was

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 35
 The laboratory’s pump capa­city limited the
0 100 maximum unit discharge to q = 50 m2/s.
Four extra independent repeatability tests

Air concentration (%)

–6 80
were done: two each at qp = 30 m2/s and
–12 60 qp = 40 m2/s. This gave two sets of three tests
hp (m)

to analyse the repeatability of the tests.

–18 40

–24
–30
0 1 2 3
Hp (m)
Minimum pressure Cavitation threshold (–7 m)
Air % Air % threshold (8%)
Figure 10 P
 rototype pressure and air concentration of whole splitter – unaerated model
measured at the positions where cavita-
tion was expected. This was compared to
Chanson’s (1992) recommendation that at
least 8% air content within the flow would
mitigate the damaging potential of cavita-
tion. An intrusive conductive needle probe
was used to measure the air concentration
of the flow. The diameter of the probe’s
conductive tip is Ø0.1 mm. The probe oper-
ates by measuring the electric conductivity
of air and water, and returning the acquired
voltage to a data logger. The data was
logged using Thermo Needle Probe (TNP)
software. Both the probe and the software
were supplied and calibrated by German-
based HZDR Innovation. From Calitz’s
(2015) previous studies, a 1 minute duration
at a frequency of 10 kHz was sufficient to
produce a representative data set. The air
concentration was measured at three posi-
tions close to the surface of the splitters in
order to determine a representative Air%
value. These three positions were:
■■ At the centre of the downstream end
of the splitter, 5 mm from the splitter’s
surface
■■ At the centre of the side face of the
splitter, 5 mm from the splitter’s surface
20
0
4 5 6 7 8
Absolute zero (–10.2 m)
■■ At the centre of the bottom face of
the splitter, 10 mm from the splitter’s
surface.
Air discharge of air vents on splitters
The air discharge of the air vents was calcu-
lated from air velocity measurements taken
within the aeration gallery of both the aerated
models. These velocity measurements were
taken using a Lutron hot-wire anemometer
with an accuracy of ±1%. Data was recorded
by hand for 2 minutes at 5 second intervals,
before the average air velocity per test was
used to calculate the air discharge per split-
ter. The air discharge per air vent could not
be determined, as it was impossible to take
measurements within the splitter itself.
Test conditions
The tests were divided between two measur-
ing splitters to accommodate the number
of pressure sensors within the splitter, as
explained above. Thus, six rounds of tests
were done, one for each measuring splitter
per model. In each round, ten tests were per-
formed starting, at qp = 5 m2/s and ending
at qp = 50 m2/s, increasing in steps of 5 m2/s
(where qp is the prototype unit discharge).
Discussion of results
All values are prototype values unless
­otherwise indicated.
Drowning of unaerated
Roberts splitters
A common theme with the results of the
unaerated model tests was the severe
change in conditions from spillway
heads less than the design head of 6.7 m
(q = 40 m2/s), to spillway heads equal
to and greater than 6.7 m. The pressure
around the entire splitter dropped consid-
erably to below the cavitation threshold of
–7 m atmospheric, and the air concentra-
tion decreased to 0%. Figure 10 shows the
prototype pressure and air concentration of
the entire splitter. The probable minimum
pressure of all 25 pressure sensors and
the average air concentration of all three
measured positions were calculated to
provide this visual illustration of the drop
in performance of the splitters.
From Figure 10 it is clear that the prob-
able minimum pressure of the entire splitter
drops to the assumed cavitation threshold
of –7 m atmospheric (Chadwick et al 2013),
and that the air concentration drops to
below the 8% needed to alleviate cavitation
(Chanson 1992). (It should be noted that
the pressures recorded in the model and
converted to prototype which are lower than
prototype cavitation pressure will remain
at cavitation pressure in prototype.) It was
concluded that at unit discharges greater
than 35 m2/s, i.e. at spillway heads equal to

Air pocket

Air pocket
(a) Top view (b) Side view

 hotographs showing normal operation of Roberts splitters at H = 6.1 m (q = 35 m2/s)

Figure 11 P

36 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering

No air
 hotographs showing drowned operation of Roberts splitters at H = 6.7 m (q = 40 m2/s)
Figure 12 P
Flow
d 0.8
θ L pro
j
Figure 13 D
 efinition of d and Lproj on a cross-
section of a splitter
and higher than the design head of the ogee
spillway profile of 6.7 m (q = 40 m2/s), the
cavitation risk of unaerated Roberts splitters
is so high it can be deemed inevitable that
cavitation will occur. This was due to the
drowning of the splitters, meaning the air
pockets that form in the boil downstream of
the splitters were being washed downstream
by the flood over the spillway. It should be
explained that it was possible to measure
100% air concentration due to the air pockets
that formed downstream of the splitters.
Especially during low flows, no water would
enter these air pockets at all. But, as the
flow increased, water would occasionally
enter these pockets, up until the moment
the splitters became drowned at H = 6.7 m
(q = 40 m2/s), when the entire air pocket was
washed downstream by the strong current
over the splitters.
Jordaan (1989) stated that the energy-
dissipating performance of Roberts splitters
drops when the splitters become drowned
at the critical head of 1.2 times the design
head. The unaerated model tests found that
this critical head was reached at exactly the
design head of 6.7 m (q = 40 m2/s) and not
at 1.2 times the design head. During the
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
(a) Top view
1.2
1.0
d / Lproj
0.6
0.4
0.2
0 0
1 2 3 4
d/Lproj
Figure 14 Relationship of depth of flow and spillway head to the projected splitter height
development of the splitter system, Roberts
(1943) not only prescribed the critical head
limit of 1.2HD, but also limited the use
of unaerated Roberts splitters to spillway
heads of only 3.0 m. Figures 11(a) and (b),
and Figures 12(a) and (b) show photographs
of unaerated model tests at prototype spill-
way heads (Hp) of 6.1 m (q = 35 m2/s) and
6.7 m (q = 40 m2/s) respectively – note that
the design head of the ogee profile of the
spillway model was 6.7 m (q = 40 m2/s).
An air pocket can clearly be seen in
the boil downstream of the splitters in
Figure 11. The results indicate that this air
pocket greatly alleviated severe negative
pressures on the splitters. This was due to
the compressibility of air in comparison to
water. If air was present within the flow, it
absorbed the extreme pressure fluctuations
caused by the turbulent flow around the
splitters by compressing and expanding
accordingly. In contrast, if the air pocket
was absent, as is the case with the drowned
splitters, the extreme pressure fluctuations
were transferred to the surface of the split-
ters by the incompressible water.
The study found that the drowning of
the splitters commenced at a critical flow
depth d approaching the splitters, and
No air
(b) Side view
3.0
2.5
2.0
H / Lproj
1.5
1.0
0.5
5 6 7 8
Hp (m)
H/Lproj Drowned limit
that the relationship between d and the
projected height of the splitters played a
role. This is also evident from Figure 12.
The projected height of the splitters was
defined as Lproj and is related to the length
of the splitters L and the downstream
slope of the spillway θ, and can be calcu-
lated using Equation 1 (see Figure 13).
Lproj = L × sin (θ)(1)
In the case of the studied hydraulic model,
the prototype spillway slope was 0.75:1
(H:V), equal to a slope of θ = 53.1°, and the
prototype length of the splitters was 3.425.
This gave a projected splitter height of
Lproj = 2.740 m.
Using the Bernoulli energy equations,
the depth of flow was calculated from the
spillway head H, the spillway unit discharge
q and the height of the splitters below the
spillway crest P (Equation 2).
(q/d)2
H+P=d+ (2)
2∙g
From the depth of flow, the relation-
ship of d/Lproj was obtained and plotted
in Figure 14 against the spillway head
37
to determine when the splitters would
become drowned. The plot of Hp/Lproj
is included for convenience. Note that
the usage of the relationship of Hp/Lproj
to determine the point at which Roberts
splitters will start to drown is subject to
the discharge coefficient of the specific
spillway. Using the observation that the
splitter became drowned at the prototype
design spillway head of 6.7 m (q = 40 m2/s),
from Figure 14 it can be concluded that
for high-design spillway heads of around
6.7 m (q = 40 m2/s), Roberts splitters could
become drowned at a point where H/Lproj is
equal to 2.4, and d/Lproj is equal to 0.84.
Thus, if the flow was deep enough to
effectively flow over the splitters without Figure 15 P hotograph showing lack of air downstream of splitters of the unaerated model – H = 6.7 m
projecting away, it would flush the air
pocket downstream of the step, causing
severe pressure conditions around the step, 0
as no air is present to absorb the negative –2
pressures. Figure 15 shows a close-up pho-
–4
tograph of the splitters at the design head
hp (m)

of 6.7 m (q = 40 m2/s), showing the lack of –6

an air pocket downstream of the splitters. –8

–10
Comparison of local pressures
–12
and air concentration 0 1 2 3 4 5 6 7 8
Figure 16 contains the comparison of Hp (m)
prototype minimum pressures of the whole Unaerated First aerated Cavitation threshold
splitter, while Figure 17 shows the air Second aerated Absolute zero
concentration for each model. The models
behaved similarly for heads up to 6.1 m Figure 16 C
 omparison of splitter minimum pressures of all models
(q = 35 m2/s) in terms of pressure and air
concentration. However, at the design head
of 6.7 m (q = 40 m2/s) and above, the split- 100
ters became drowned (as discussed above)
Air concentration (%)

and the unaerated splitters showed a severe
drop in pressure and air concentration.
The normal air pocket found in the boil
immediately downstream of the splitters
was flushed downstream of the step, and
the characteristic air entrainment mecha-
nism of Robert splitters was not as effective
in this state as intended by Roberts (1943).
Pressures close to high vacuum conditions
(–10.2 m) were measured at the side face,
along with 0% air concentration, leading
to the conclusion that cavitation would
inevitably occur.
The air vents of the aerated models
provided much needed air supply to the
splitters in this drowned state. The air
pocket was continuously re-established and
the pressure conditions improved consider-
ably, as evident from Figure 16. At spillway
heads equal to and greater than the design
spillway head of 6.7 m (q = 40 m2/s), the
change in prototype pressure is clearly evi-
dent and beneficial. The air concentration
80
60
40
20
0
0 1 2 3 4 5 6 7 8
Hp (m)
Unaerated air percentage First aerated air percentage
Second aerated air percentage Air percentage threshold (8%)
Figure 17 C
 omparison of splitter average air concentration for all models
still decreased for the increasing spillway concentration of the second aerated model
head, but did not drop below 40%. In only decreased slightly compared to that of the
one location was a prototype pressure first aerated model. It was concluded that
recorded that was less than the cavitation the air concentration was not necessarily
limit of –7 m, as prescribed by Chadwick as dependent on the size of the air vents,
et al (2013), and, according to Chanson’s but rather on the presence of the air pocket
(1992) findings, this is acceptable, given downstream of a splitter.
that the air concentration in the flow at Figures 18 and 19 show 2D contour
this point is greater than 8%. Referring to plots of the minimum measured prototype
Figure 17, it should be noted that the air pressure on the side, end and bottom

38 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 Unaerated model First aerated model Second aerated model

Side face Side face Side face

60 60 60

40 40 40

y (mm)
y (mm)
y (mm)
20 20 20

Side face
No pressure sensors No pressure sensors No pressure sensors

0 0 0
0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160
x (mm) x (mm) x (mm) 0

End face End face End face

60 60 60 –5

40 40 40
–10

y (mm)
y (mm)
y (mm)

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
20 20 20

End face
–15
0 0 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120
x (mm) x (mm) x (mm)
–20
Prototype pressure head (m)

Bottom face Bottom face Bottom face

100 100 100 –25

80 80 80
–30
60 60 60

y (mm)
y (mm)
y (mm)

40 40 40

Bottom face
20 20 20

0 0 0
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
x (mm) x (mm) x (mm)

39
Figure 18 P
 ressure distribution on the splitter for all models at Hp = 6.7 m
40
Unaerated model First aerated model Second aerated model

Side face Side face Side face

60 60 60

40 40 40

y (mm)
y (mm)
y (mm)
20 20 20

Side face
No pressure sensors No pressure sensors No pressure sensors

0 0 0
0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160
x (mm) x (mm) x (mm) 0

End face End face End face

60 60 60 –5

40 40 40
–10

y (mm)
y (mm)
y (mm)
20 20 20

End face
–15
0 0 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120
x (mm) x (mm) x (mm)
–20
Prototype pressure head (m)

Bottom face Bottom face Bottom face

100 100 100 –25

80 80 80
–30
60 60 60

y (mm)
y (mm)
y (mm)

40 40 40

Bottom face
20 20 20

0 0 0
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
x (mm) x (mm) x (mm)

Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Figure 19 P
 ressure distribution on the splitter for all models at Hp = 7.6 m
faces of the model splitter for the design
spillway head of 6.7 m (q = 40 m2/s) and
the maximum tested spillway head of 7.6 m
(q = 50 m2/s). For the definitions of the
faces of the splitter, refer to Figure 7. The
side bottom face was excluded from these
plots, as the four sensors F1 to F4 provided
too little data. The blank zone towards the
bottom right of the side face indicates that
no sensors were placed in that area, and
as such, no data existed to plot there. The
dimensions of the splitter faces in Figures
18 and 19 are to model dimensions.
The plots clearly indicate that the loca-
tions of the most severe negative pressures
on the unaerated model were on the side
face, near the top and close to the spillway
surface. The most severe negative pressure
on the end face was measured in the centre,
and the most severe negative pressure on
the bottom face was measured near the top
at the joint with the end face. This justifies
the design and placement of the air vents.
Regarding the pressure distributions
of the design head of 6.7 m (q = 40 m2/s)
in Figure 18, it was noted that the severe
pressure situation of the unaerated model
was completely alleviated by both the
aerated models. At the maximum head of
7.6 m (q = 50 m2/s), the absolute value of
the minimum measured pressure on the
splitter is similar for the first and second
aerated models, as evident from Figure 16.
However, the spread and distribution
thereof are different. Regarding all splitter
faces of the first aerated model in Figure 19,
note the large spread of prototype pressure
equal to or less than the –7 m cavitation
0
hp (m)
–2
–4
–6
–8
0 1 2 3
First aerated: Minimum
Second aerated: Minimum
Figure 21 Prototype pressure at the main air vents of both aerated models
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
10 0.10
8 0.08
Qinflow (m3/s)
Qair (m3/s)
6 0.06
4 0.04
2 0.02
0 0
1 2 3 4 5 6 7 8
Hp (m)
Air discharge: First aerated model Air discharge: Second aerated model
Water inflow: First aerated model Water inflow: Second aerated model
Figure 20 P
 rototype air discharge and water inflow per splitter for both aerated models
limit set by Chadwick et al (2013) (the red As the movement of air and water
areas on the contour plots). This is greatly through the air vents is caused by the
reduced in the second aerated model to a pressure differential between the outside
single zone on the bottom face. As men- and inside of the air vents, and the pres-
tioned, this was acceptable, given that the sure on the inside of the vents was close to
air concentration in the flow at this point is atmospheric, the measured pressure close
greater than 8% (Chanson 1992). to the air vent positions was analysed to
investigate the flow of air (and possibly
Discussion on the performance water) through the air vents.
of the aeration system The prototype pressures at the main air
The main considerations in estimating the vents were interpolated from the values of
performance of the proposed aeration system pressure sensors A2, A3, B2 and B3 (see
were air discharge and water inflow per Figure 9 for the positions of the sensors),
splitter. Figure 20 shows the prototype air while the pressures of the side air vents
discharge (Qair) and the water inflow (Qinflow) were interpolated from sensors E4, E5 and
per splitter for all tested spillway heads. It E6. By taking the minimum, average and
is evident that the decrease in air discharge maximum of these interpolated sets of
at heads greater than 5.0 m (q = 25 m2/s), data, Figures 21 and 22 were produced.
and more prominently at heads greater than In Figures 21 and 22 the negative pres-
6.1 m (q = 35 m2/s), is linked to the increase sures relate to air being discharged from
in water inflow at the same heads. the air vents, and the positive pressure
4 5 6 7 8
Hp (m)
First aerated: Average First aerated: Maximum
Second aerated: Average Second aerated: Maximum
41
 1
–1
–2
hp (m)
–3
–4
–5
–6
0 1 2 3
First aerated: Minimum
Second aerated: Minimum
Figure 22 Prototype pressure at the side air vents of both aerated models
relates to water that flowed into the vents.
It must be noted that the minimum and
maximum pressure values in these figures
represent pressure peaks. There were
several positive peaks and several negative
peaks. Each of these peaks would induce
a burst of air discharge or a burst of water
inflow. Air and water never continuously
or steadily flowed through the air vents.
However, the average pressures (which
are negative in all cases) imply that, over
extended periods, air will be discharged
out of the vents rather than into them.
The large increase in air discharge of
the second aerated model from that of the
first aerated model at heads between 1.9 m
(q = 5 m2/s) and 7.1 m (q = 45 m2/s) can be
explained by the similarity of the pressure
results between the two aerated models
at the locations of the air vents. If the
pressures at the air vents were to be equal,
then the larger cross-sectional area of the
Ø23 mm vents would produce a greater
volume of air per given time. Referring to
the higher heads of 7.1 m (q = 45 m2/s) and
7.6 m (q = 50 m2/s) within Figures 21 and
22, the pressure at the air vents of the sec-
ond aerated model increased from that of
the first aerated model, causing a decrease
in air demand, and thus air discharge.
The larger air vents still resulted in an
increase in air discharge, as can be seen in
Figure 20, but the decreased demand meant
that the difference between the two aerated
models is not as great as is the case with
the heads less than 7.1 m (q = 45 m2/s).
Referring to Figure 20, small amounts of
water only started sporadically flowing into
42
4 5
Hp (m)
First aerated: Average
Second aerated: Average
the air vents at spillway heads equal to and
greater than 6.7 m (q = 40 m2/s), the same
head as when the splitters became drowned.
As mentioned, the inflow of water is due to
peaks of positive pressure at the air vents.
From observations during the tests it was
found that almost no water flowed into the
side vents. This is partly due to the size of
the side vents being half of that of the main
vents, but mostly due to absence of notable
positive pressure at the side vents, as can be
seen in Figure 22. During the aerated tests,
the sides of the splitters produced surpris-
ingly stable and predictable pressure results,
especially considering that during the
unaerated model tests the pressures on the
side were the most severe of all tests.
Conclusions
The conclusions from the hydraulic model
study can be summarised as follows, where
all values are prototype values, unless
otherwise stated:
■■ It was found that the splitters became
drowned at the design spillway head of
6.7 m (q = 40 m2/s).
■■ A system of unaerated Roberts split-
ters should not drown under these
circumstances:
■■ d/Lproj < 0.84
■■ H/Lproj < 2.4
W here d is the depth of flow approach-
ing the splitters, H is the spillway head
and Lproj is the projected height of the
splitters in the direction of the flow
(see section above titled “Discussion
of results”).
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
6 7 8
First aerated: Maximum
Second aerated: Maximum
■■ These drowned conditions would
inevitably lead to cavitation of unaer-
ated splitters. The minimum pressure
was measured at –10.2 m and the air
concentration was 0%.
■■ The introduction of artificial aeration
through air vents greatly alleviated
these cavitation risks.
■■ The following favourable combinations
of pressure and air concentration were
measured for each model at the design
spillway head of 6.7 m (q = 40 m2/s):
■■ First aerated model: –3.7 m pressure
and 66% air concentration
■■ Second aerated model: –2.3 m pres-
sure and 55% air concentration
In reply to Jordaan’s (1989) statement,
the need for aeration to Roberts split-
ters, especially at high spillway heads, is
supported by these results.
■■ With the increase in air vent size of the
second aerated model, the severe nega-
tive pressure conditions were signifi-
cantly improved at the maximum tested
spillway head of 7.6 m (q = 50 m2/s – see
Figures 18 and 19).
■■ The observed decrease in air discharge
at unit discharges higher than 25 m2/s
was due to higher pressure volatility
and more regular and higher positive
pressure peaks at the air vent positions.
This resulted in water inflow into the
air vents that needed to be continually
drained. This, however, did not influ-
ence the performance of the model’s
aeration gallery. The peak water drain-
age per splitter was 0.092 m3/s (second
aerated model tests).
■■ The maximum measured unit discharge
that was safely passed over the spillway
splitters was 50 m2/s (H = 7.6 m), which
is significantly more than the 12 m2/s
(H = 3 m) limit set by Roberts (1943)
(see section above titled “Limitations to
Roberts splitters”) and the limit of 35
m2/s (H = 6.1 m2/s) as measured in this
study for unaerated splitters.
■■ The unit discharge of 50 m2/s was
limited due to the laboratory’s pump
capacity. The tests showed that Roberts
splitters could possibly be used at higher
unit discharges, but this was not evalu-
ated during the study.
In closing, it was found that the main
objective of the study was successfully
achieved, as aeration of Roberts splitters,
through an internal gallery, had a satisfac-
tory effect on the local negative pressures
around the splitters, and sufficiently allevi-
ated the cavitation risk at prototype unit
discharges of up to 50 m2/s (H = 7.6 m).
Recommendations for
further work
The following summary can be used as
starting points for further investigation
into establishing practical design guidelines
for aerated Roberts splitters at spillway
heads greater than the 3 m limit as set by
Roberts (1943):
■■ Test and evaluate a broad spectrum of
splitter configurations, based on a wide
range of design heads, up to and exceed-
ing at least 10 m. These tests should
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
primarily focus on the energy dissipa-
tion and downstream dispersive action
of the splitters before focusing on local
structural and surface effects.
■■ With the focus on the air vents, a wider
range of air vents should be investigated
before a set of guidelines can be prepared
with regard to the splitter air vents.
■■ The aeration gallery can be optimised
and compared to the current bottom
vent (see section above titled “Existing
aeration of Roberts splitters”) with
regard to air supply and avoiding poten-
tial drainage problems that may arise
from water inflow into the air vents.
References
Back, P A A, Frey, J P & Johnson, G 1973. P. K. le
Roux Dam spillway design and energy dissipation.
Proceedings, 11th ICOLD Congress, Madrid, Spain,
Vol. Q. 41, R. 76, No. II, pp 1439–1468.
Bosman, E & Basson, G R 2012. Investigation of
unsteady flow conditions at dam bottom outlet
works due to air entrainment during gate closure:
Physical modelling. WRC Report TT528-12,
Pretoria: Water Research Commission.
Calitz, G 2014. The effect of aeration on local negative
pressures of Roberts splitters. Final year research
dissertation, Stellenbosch University.
Calitz, G & Basson, G R 2015. The design of Roberts
splitters for energy dissipation at dam spillways.
Design and Construction of Hydraulic Structures
Course, Stellenbosch University, September 2015.
Calitz, J A 2015. Investigation of air concentration and
pressures of a stepped spillway equipped with a crest
pier. MEng dissertation, Stellenbosch University.
Chadwick, A, Morfett, J & Borthwick, M 2013.
Hydraulics in Civil and Environmental Engineering,
5th ed. Boca Raton, FL: Taylor & Francis.
Chanson, H 1992. Air entrainment in chutes and
spillways. Report CE133, Brisbane, Australia:
University of Queensland, Department of Civil
Engineering.
Jordaan, J M 1989. The Roberts splitter: Fifty years on.
The Civil Engineer in South Africa, 31(10): 319–321.
Langa, M C 2015. Dam spillway energy dissipation by
Roberts splitters: Study of cavitation and aeration
requirement. Final year research dissertation,
Stellenbosch University.
Mason, P J 1983. Energy dissipating crest splitters for
concrete dams. International Water Power & Dam
Construction, 35(11): 37–40.
Roberts, C P R 1980. Hydraulic design of dams. Report.
Pretoria: Department of Water Affairs, Forestry and
Environmental Conservation, Division of Special
Tasks.
Roberts, D F 1943. The dissipation of the energy of a
flood passing over a high dam. Proceedings of the
South African Society of Civil Engineers, XLI(1):
48–92.
Roberts, P R 1977. Energy dissipation by dam crest
splitters. The Civil Engineer in South Africa, 19(11):
263–264.
Robertson, G K 2014. Labyrinth weir hydraulics:
Validation of CFD modelling. MEng dissertation,
Stellenbosch University.
USBR (United States Bureau of Reclamation) 1987.
Design of Small Dams, 3rd ed. Washington, DC: US
Department of the Interior: USBR.
Vanderkloof Dam 2014. About Vanderkloof Dam.
Available at: http://www.vanderkloofdam.co.za
(accessed on 24 October 2014).
43
TECHNICAL PAPER Comparative analysis of
Journal of the South African
Institution of Civil Engineering methods of pile-bearing
capacity evaluation using
ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 44–55, Paper 1410

IFEANYICHUKWU OBETA is a lecturer in the

Department of Civil Engineering at the
CPT logs from tropical soils
University of Nigeria, Nsukka. He holds a
Master’s degree in soil mechanics and
foundation engineering from the same I N Obeta, M E Onyia, D A Obiekwe
university, and specialises in geotechnical
engineering. He also runs a financial and
engineering consulting firm.
This study presents the results of a comparative analysis of the performance of eight direct cone
Contact details:
Department of Civil Engineering
penetration test (CPT) methods in predicting the ultimate load-carrying capacity of a 300 mm
University of Nigeria diameter end-bearing pile, evaluated at defined soil depths using CPT logs obtained for various
Nsukka locations in the south-east and south-south regions of Nigeria. The methods used are the
Nigeria
Schmertmann (1978), LCPC (Bustamante & Gianeselli 1982 – Laboratoire Central des Ponts et
T: +234 80 6110 9704
E: ifeanyi.obeta@unn.edu.ng / obetaifeanyi@gmail.com Chausees), De Ruiter and Beringen (1979), Tumay and Fakhroo (1982), Price and Wardle (1982),
Philipponnat (1980), Aoki and De Alencar (1975), and the Penpile methods (Clisby et al 1978). The
DR MICHAEL ONYIA is a senior lecturer in the results of univariate analysis of variance indicated that the interactions between location and
Department of Civil Engineering at the treatment (methods of pile capacity prediction), depth and treatment, and location and depth
University of Nigeria, Nsukka. He holds a
were statistically significant for cohesionless soils, but that the interaction between depth and
doctorate in structural engineering from the
same university and specialises in structural treatment was not statistically significant for cohesive soils. Also, post-hoc tests (Least square
and geotechnical engineering. He is also a difference and Bonferroni methods) showed that the LCPC and Philipponnat methods are best
member of various professional bodies. suited for cohesionless soils, while the LCPC, Tumay and Fakhroo (1982) and De Ruiter and
Beringen (1979) methods are recommended for cohesive soils.
Contact details:
Department of Civil Engineering
University of Nigeria
Nsukka INTRODUCTION problems and limitations associated with the
Nigeria
One of the main steps for the safe and SPT methods are due to the fact that it does
T: +234 80 3382 1550
E: ebieonyia64@yahoo.com economic design of foundations is the deter- not provide continuous data, and therefore
mination of ultimate bearing capacity. The important information on weak seams may
DAMIAN OBIEKWE is a graduate engineer four approaches generally used for bearing be missed. It has limited applicability to
from the Department of Civil Engineering, capacity determination are static analysis, cohesive soils, gravels, cobbles and boulders.
University of Nigeria, Nsukka. He
in-situ testing methods, full-scale loading Its progress is also slower than other in-place
graduated from the said institution in
2013, and is currently a site engineer at JLG tests and using presumed values recom- tests, because of incremental drilling, testing
Development Nigeria Ltd, where he mended by codes and handbooks. Among and sample retrieval.
supervises various engineering projects. these, the in-situ methods have in recent In this regard, the authors are seeking
years become more popular, and this is to appraise the performance of the eight
Contact details:
JLG Nigerian Ltd attributed to the rapid development of in-situ generally adopted methods of estimating
Port Harcourt test methods, testing instruments, improved pile capacity from CPT logs on tropical
Nigeria understanding of soil behaviour and the soils of Group B subgroup (b) and Group C
T: +234 81 3870 3276
subsequent insight into the limitations and subgroups (b) and (c) (see Table 1). This is
E: damianobiekwe@gmail.com
inadequacies of some conventional labora- aimed at finding out if some or all of these
tory testing methods (Eslami & Gholami eight methods are applicable to tropical soils
2006; Shooshpaasha et al 2013). Among the since they were originally developed for soils
in-situ methods, the CPT method is simple, in temperate regions, but are often used for
fast and relatively economical, and supplies tropical soils as there are no documented
continuous records with depth. Its results methods solely developed for tropical soils.
are interpretable on both an empirical and
analytical basis, and a variety of sensors can
be incorporated in its use. When compared BACKGROUND
to the standard penetration test (SPT)
method, the CPT method is more popular Tropical and temperate soils
owing to the many problems and limita- Tropical soils vary in type and composi-
tions associated with the performance and tion, as is the case with soils in the temper-
Keywords: comparative analysis, cone penetration test, pile capacity, interpretation of the results obtained from ate regions of the world, but tropical soils
univariate analysis of variance, post-hoc tests the SPT (Eslami & Gholami 2006). Such possess certain peculiarities for which their

Obeta IN, Onyia ME, Obiekwe DA. Comparative analysis of methods of pile-bearing capacity evaluation using CPT logs from tropical soils.
44 J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1410, 12 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a5
Table 1 Classification of residual soils
Group Common
Means of Comments on likely engineering
pedagogical Examples
Major group Sub-group identification properties and behaviour
names

Highly weathered
This is a very large group of soils (including
rocks from acidic
(a) Strong macro- the saprolites) where behaviour (especially
Kaolinitic saprolites or intermediate Visual inspection
structure influence in slopes) is dominated by the influence of
igneous or
discontinuities, fissures, etc.
sedimentary rocks

These soils are essentially homogenous and

Group A
Completely form a tidy group much more amenable
Soils with Visual inspection
weathered rock to rigorous analysis than group (a) above.
a strong (b) Strong micro-structure and evaluation
Oxisols formed from Identification of nature and role of bonding
mineralogical influence of sensitivity,
igneous or (from relic primary bonds to weak secondary
influence liquidity index, etc
sedimentary rocks bonds) important to understanding
behaviour.

Little or no
Soils formed from This is a relatively minor sub-group. Likely
(c) Little or no structural sensitivity,
very homogenous to behave similarly to moderately over-
influence uniform
rocks consolidated soils.
appearance

■■ Black cotton soils Black cotton soils

■■ Black soils and many similar Dark colour (grey These are normally problem soils found
Group B (a) Smectite
■■ Tropical black dark-coloured to black) and high in flat and low-lying areas, having low
Soils strongly (montmorillonite)
earths soils formed in plasticity suggest strength, high compressibility, and high
influenced by group
■■ Grumusols poorly drained soils of this group swelling and shrinkage characteristics.
normal clay ■■ Vertisols conditions
minerals
(b) Other clay minerals Likely to be a very minor sub-group.

Soils weathered Position on Characterised by very high natural water

■■ Volcanic ash soils
from volcanic ash plasticity chart content and Atterberg’s limits. Engineering
■■ Andosols/
(a) Allophane sub-group in the wet tropics and irreversible properties are generally good, though
andisols
and temperate changes on in some cases high sensitivity may make
■■ Andepts
climates drying earthworks difficult.

Group C Soils often Reddish colour, These are generally fine-grained soils of
■■ Tropical red clays
Soils strongly derived from well drained low to medium plasticity and low activity.
■■ Latosols
influenced by (b) Halloysite sub-group volcanic material, topography and Engineering properties are generally good
■■ Oxisols
clay minerals especially tropical volcanic origin are (there is often some overlap between
■■ Feralsols
essentially red clays useful indicators halloysite and allophane clays).
found only in
residual soils Non-plastic or
This is a very wide, poorly defined group
low plasticity
■■ Lateritic soils Laterites or ranging from silty clay to coarse sand and
(c) Sesquioxides – materials,
■■ Laterites possibly some red gravel. Behaviour ranges from low plasticity
gibbsite, goethite, generally
■■ Ferralitic soils clays referred to silty clay to gravel. These materials are the
haematite of granular
■■ Duricrusts as lateritic clays end products of a very large weathering
or nodular
process.
appearance
Source Wesley (2009)

location is a contributing factor. They are
located between the Tropic of Cancer and
the Tropic of Capricorn (Punke 2014), with
a large variety of them being residual soils.
They are residual as a result of intense
chemical weathering initiated by climatic
circumstances that break down clay miner-
als into hydrated oxides of aluminium and
iron, whereas clay is fairly stable in temper-
ate regions. Also, in the warm and humid
climates typical of the tropics, the time
required to chemically alter a rock mate-
rial is considerably less than in temperate
climates (Morin & Todor 1976). Residual
soils are different from the sedimentary
soils of the temperate regions which are
formed through a depositional process
wherein soils are deposited in a marine/
lake environment. Two important factors
that lead to a degree of homogeneity and
predictability of sedimentary soils that are
absent from residual soils are:
■■ The sorting process that takes place dur-
ing erosion, transportation and deposi-
tion of sedimentary soils, which tend to
produce homogeneous deposits, and
■■ The stress history, which is a prominent
factor in determining the behavioural
characteristics of sedimentary soils, and
leads to the convenient division of these
soils into normal and over-consolidated
materials, unlike residual soils for
which mineralogy, and not stress his-
tory (a concept which has not much if
any relevance), is the important factor
(Wesley 2009; Nnadi 1987). The general
climatic requirements for the formation
of residual soils are an average annual
rainfall of at least 1 200 mm and a
daily temperature in excess of 25°C
(Gogo-abite 2005). The classification
of residual soils is presented in Table 1,
while other peculiar tropical soils, such
as laterites, tropical red clays, peats and
black cotton soils are discussed in the
following paragraphs.
Laterites
Rahardjo et al (2004) described laterites as
a product of the in-situ weathering of igne-
ous, sedimentary and metamorphic rocks
under unsaturated conditions. According
to Information Technology Associates
USA (ITA USA 2011), “Chemical and

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 45
microscopic investigations have shown that
lateritic clay differs from those commonly
found in temperate regions. It does not
contain the hydrous silicate of alumina,
but is a mechanical mixture of fine grains
of quartz with minute scales of hydrates of
alumina. Lateritic clay is easily soluble in
acid while clay is not, and after treating lat-
erite with acids the alumina and iron leave
the silica as a residue in the form of quartz.
The great abundance of alumina in some
varieties of laterite is a consequence of the
removal of the fine particles of gibbsite,
hydrargillite and diaspore from the quartz
by the action of gentle currents of water.”
A summary of the geotechnical proper-
ties of laterites is given in Table 2.
Tropical red clays
Tropical red clays are formed by the
weathering of various types of rocks which
range from granite, volcanic lavas and ashes
(Anon 1990). The colour of the soils could be
brown, reddish brown or pure red, and they
contain iron sesquioxides and halloysites,
kaolinite and/or allophane clay minerals
(Culshaw et al 1992). A feature of these red
clays is that they possess enough cohesion
to enable an unsupported block of the soil
to be trimmed or carved in-situ, and this is
as a result of their negative pore pressures
which increase their effective strength and
enhance their ability to stand unsupported.
Geotechnically they are unusual in that,
despite having high clay content (over 50%),
intact specimens are friable and comprise
an open, weakly bonded structure of silt-
sized clay particles resulting in high void
ratios and low densities. They dry rapidly on
exposure and are prone to stress-sensitive
collapse (Culshaw et al 1992; Hobbs et al
1992). Considering the subject of sample
disturbance which arises as a result of the
various methods of sampling, Terzaghi and
Peck (1967) reported that the value of over-
consolidation ratios (OCRs) of temperate
soils (using traditional sampling methods),
estimated via the method of Schmertmann
(1953), lies between 0.3 and 0.7. However,
Culshaw et al (1992) observed that the tropi-
cal red clays of Kenya have OCRs greater
than 1, and as such they concluded that the
method of Schmertmann cannot be used
to quantify the effect of sample disturbance
for tropical soils. This they attributed to the
variable bonding strength of the bonded red
clays, and the extent to which this bonding
strength is reduced by sampling cannot be
ascertained without comparative experimen-
tal data. In addition, it has been observed by
46
Table 2 Geotechnical properties of laterites
Geotechnical
Behavioural trend
property
Density increases with depth in the laterite deposits. The upper layers are more
porous and possess a high void ratio as a result of weathering and leaching of soil
minerals, while the lower layers have a lower porosity and void ratio, as a result
Density
of finer particles resulting from the breakdown of the upper granular structure
by the removal of the sesquioxides cementing agents (Rahardjo et al 2004;
Townsend et al 1973).
The compressibility of laterite soils is generally low and its compaction results in
Compressibility
improved unit weight, void ratio and compression index (Ogunsanwo 1990).
The permeability reduces with increase in depth, as a result of the leaching of
fines combined with the effect of weathering, which fills up the pores in the lower
Permeability
deposits. Compacted laterites also show a reduction in permeability as the degree
of saturation increases (Nnadi 1987).
The factors that affect the shear strength of laterites are degree of weathering,
mineralogical and chemical composition, and water content. The shear strength
parameters generally increase with depth as weathering decreases with depth
Shear strength (Rahardjo et al 2004). Increase in kaolinite content increases cohesion, increase in
sesquioxide content increases the internal angle of friction, but increase in moisture
leading to inundation results in drastic reduction of stability capacity and leads to
shear failure (Gogo-abite, 2005).
Wesley (2009) that tropical red clays of Java, roots and root exudates is therefore
Indonesia, when plotted on a conventional much greater in tropical than in temper-
e-logP graph, behave as moderately over- ate peat deposits. Hence, the rubbing test
consolidated soil, but when the consolidation and examination of liquid extracted from
behaviour of the same soil is plotted on a tropical peat lead to incorrect charac-
linear compression vs P graph, the curves terisation of texture, which is often fibric
generated are reasonably close to linear and because of woody components.
the evidence of yield stress disappears. He ■■ Existing classification schemes for
also reported that sample remoulding has temperate and boreal peats are based on
no effect on the compression behaviour (e vs selected characteristics for specific uses
logP plot) of the red clays, because they exist in the fields of agriculture, engineering,
naturally in a dense, unstructured state that energy, etc, rather than having a generic
is close to their liquid limit. approach. This results in a lack of cor-
relation between field observations and
Tropical peats laboratory test results, because field
Tropical peats are organic soils deposited investigations are often regarded as less
in mires of the tropics and subtropics lying important than laboratory results, and
between latitudes 35˚N and 35˚S, including also limits the interpretation of data
those at high altitudes (Andriesse 1988). collected in the field.
They possess ash contents in the range ■■ Classifications of organic soil for agri-
of 0–55% (Wust et al 2003). A study of cultural purposes are based on a control
the tropical peat deposits of Tasek Bera, section. Hence, a 5 m thick peat deposit
Malaysia, by Wust et al (2003) showed that would be classified according to the
the soil could not fit into any of the classifi- upper 50 cm section of the profile, ignor-
cation systems provided for peats found in ing the nature and origin of the underly-
the temperate regions of the world. Wust et ing deposit. A full description and char-
al (2003) gave the following reasons for this acterisation of the complete stratigraphic
peculiar behaviour of tropical peats: section is required for classification.
■■ Temperate and boreal peats are often It is therefore obvious why tropical peats
dominated by bryophytes and shrubs. are not captured in Table 1 and are thus
Root penetration is thus shallow and not considered in this work.
decomposition rates are often low. In
contrast, tropical peatlands have vari- Black cotton soils
ous tree species with roots penetrating Black cotton soils (BCS) are expansive soils
the organic deposits for several metres. (Tomlinson & Boorman 1999) that are
Rates of biomass production and primary found in the north-eastern part of Nigeria,
decomposition are high. Subsurface Cameroon, Lake Chad Basin, Sudan,
input of organic matter from decaying Ethiopia, Kenya and South Zimbabwe. The
Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Table 3 Summary of CPT pile design and prediction methods
CPT pile prediction method qb (unit end bearing capacity) f (unit shaft friction)

qc1 + qc2 In clay:

qb = ≤ 15 MPa
2 f = kc fs ≤ 120 Kpa kc = 0.2 – 1.25

In sand:
Schmertmann (1978)
8D d L
Qs = k ∑ fs As + ∑ fs As
d=0 8D d=8D

Where K depends on d/D ratio

In clay: In clay:
qb = Nc Su ≤ 15 MPa f = αSu ≤ 120 KPa

qca Α = 1 for NC clay and 0.5 for OC clay

Su = , Nc = 9, Nk = 15 to 20
Nk
In sand: In sand:

⎧
⎪ qca (compression)
De Ruiter and Beringen (1979)

⎪ 300
Similar to Schmertmann (1978) fs

⎨ qca
⎪ 400 (tension)
⎪
f = min

⎩ 120 KPa

qeq (side)
qb = kb1qeq(tip) f= ≤ max ks1 = 30 – 150
ks1
LCPC (Bustamante and Gianeselli 1982)
depending on soil type, pile type and installation
procedure
Similar to Schmertmann (1978) f = mfca ≤ 72 KPa
Tumay and Fakhroo (1982) m = 0.5 + 9.5e–0.009fca
where fca is average friction in KPa

qca(tip) qca(side)α1
qb = ≤ 15 MPa f= ≤ 120 KPa
Fb Fs2
Aoki and De Alencar (1975)
Fb depends on pile type = 1.75 for PPC driven piles α1 = 1.4 – 6 depends on soil type while Fs2 depends on
pile type = 3.5 for PPC driven piles
qb = kb2 qca(tip) f = α2 f s
Price and Wardle (1982)
where kb2 depends on pile type = 0.35 for driven piles where α2 = 0.53 for driven piles

qca(side)α3
f=
qb = kb3 qca(tip) Fs2
Philipponnat (1980)
where kb3 depends on soil type = 0.4 for sand, 0.45 for where α3 depends on pile type = 1.25 for PPC driven
silt and 0.5 for clay piles and Fs2 = 50–200 depending on soil type

⎧ 0.25qca(tip) for pile tip in clay

qb = ⎨
fs
⎩ 0.125qca(tip) for pile tip in sand
f=
Penpile (Clisby et al 1978) 1.5 + 14.47 fs
where f and fs are in MPa
qca = average qc values over a specified zone that depends on the method
Source Abu-Farsakh and Titi (2004)

soil is also found in India, Australia and the
south-west of the United States of America,
South Africa, Israel and other places
where the annual evaporation exceeds the
precipitation (Chen 1975). They contain
high amounts of montmorillonite and are
very problematic in construction (either
for building of houses or for the construc-
tion of roads) as a result of differential
settlement arising from the shrink-swell
behaviour of the soil. The swelling occurs
when polar molecules, such as water or
organic molecules, adsorb onto the tip of
the soil layers in the inter-planar space.
The expansion of the soil interlayer and
swelling is thought to be primarily influ-
enced by the type of exchangeable cations
present in the aqueous solutions that come
into contact with the clay (Luckham &
Rossi 1999; Zhou 1995). Cracks measuring
70 mm wide and up to 3 m deep have been
reported in places with high deposits of
BCS (Adeniji 1991).
MATERIALS AND METHODS
Data collection and modification
Eight CPT data (logs) in kg/cm2 compris-
ing 320 data points were collected from

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 47
MTN Nigeria telecommunications Ltd.
700
The logs were generated from results of
CPT tests carried out for the purposes of
600
designing suitable foundations for telecom-
munication towers across the Anambra, 500

Pile capacity (kN)

Imo and Delta states of Nigeria. The depth
of the logs ranged from 0.25–9.75 m below 400
ground level. The data from Anambra was
code-named ANAM 1-4, that from Imo 300

was named IMO 1-3, and the Delta data

200
was named DELTA.The soils upon which
the tests were carried out were classified 100
according to the predominant soil type,
whether cohesive or cohesionless. The 0
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
respective ultimate end-bearing pile capac-
Depth (m)
ity values Qu in KN were then computed for
Schmertmann De Ruiter LCPC Tumay & Fakhroo
each location and depth, using the methods Price & Wardle Philipponnat Penpile Aoki & De Alencar
summarised in Table 3 on page 47, while
assuming that the piles to be installed are Figure 1 P
 erformance of pile capacity methods across depths at IMO3
300 mm diameter piles. These calculations
were done for depths of 4.5, 4.75, 5.0, 5.25,
5.5, 6.5, 6.75, 7.0, 7.25 and 7.5 m for each 1 200
location, while neglecting skin friction fs,
because the parameters for its compua- 1 000
tion were not available, and by implication
considering the piles as strictly end-bearing 800
Pile capacity (kN)

piles. The unavailability of data for comput-

ing skin friction was as a result of the dutch 600
cone penetrometer models used by the
contractors. The calculated pile capacities 400
are presented in Figures 1 to 8.
200
Methods of estimating pile capacity
from the CPT test 0
Several authors have worked on the direct 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
methods of estimating pile capacity using Depth (m)
the CPT approach. The authors and their Schmertmann De Ruiter LCPC Tumay & Fakhroo
Price & Wardle Philipponnat Penpile Aoki & De Alencar
approaches are presented below:

The Schmertmann (1978) method Figure 2 Performance of pile capacity methods across depths at ANAM1
This is based on a summary of the work
on model and full-scale piles presented 1 200
by Nottingham (1975). The method is as
described in Equation 1:
1 000

qc1 + qc2
qt = (1) 800
Pile capacity (kN)

600
Where: qt is the unit tip bearing capacity
of the pile, qc1 the minimum of the average
400
cone tip resistances of zones ranging from
0.7D to 4D below the pile tip (where D is
200
the pile diameter and qc1 is determined
by the minimum path rule) and qc2 is the
0
average minimum cone tip resistances over 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
a distance 8D above the pile tip. Thus the Depth (m)
zone 8D above 0.7D–4D below the pile Schmertmann De Ruiter LCPC Tumay & Fakhroo
tip represents the failure surface, which Price & Wardle Philipponnat Penpile Aoki & De Alencar
is approximated by a logarithmic spiral.
A limitation of the Schmertmann method Figure 3 P
 erformance of pile capacity methods across depths at ANAM4

48 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 is that the skin friction cannot exceed
1 400
120 kPa and the method of determining
skin friction varies for different pile types,
1 200
especially in clay soils.
1 000
De Ruiter and Beringen (1979) method
Pile capacity (kN)

800 De Ruiter and Beringen (1979) presented

a method based on experience gained
600 in the North Sea by Fugro Consultants
International (Eslami & Fellenius 1995),
400
which is similar to Schmertmann’s method
200 for determining qt in sands, but differ in
clays. In clays qt is given by Equation 2,
0 where Nc is the bearing capacity factor, Su
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
the undrained shear strength of each soil
Depth (m)
layer and Nk the cone factor.
Schmertmann De Ruiter LCPC Tumay & Fakhroo
Price & Wardle Philipponnat Penpile Aoki & De Alencar
qc(tip)
qt = NcSu(tip) where Su(tip) = (2)
Figure 4 P
 erformance of pile capacity methods across depths at ANAM2 Nk

The LCPC (Bustamante &

600 Gianeselli 1982) method
The LCPC or the French method was
500 developed by Bustamante and Gianeselli
(1982) and is based on the analysis of
400 197 pile load tests with a variety of pile
Pile capacity (kN)

types. In this method both the unit tip

300 bearing capacity qt and the unit skin fric-
tion f of the pile are obtained from the
200 cone tip resistance qc. The sleeve friction
f s is not used (Titi et al 1999) and this
100 is unlike the other methods. The unit
tip bearing capacity qt is predicted from
0
Equation 3 for which the average qc, is
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 determined within a zone of 1.5D above
Depth (m) and 1.5D below the pile tip, and K b is an
Schmertmann De Ruiter LCPC Tumay & Fakhroo empirical bearing capacity factor varying
Price & Wardle Philipponnat Penpile Aoki & De Alencar
from 0.15 to 0.6 depending on the soil
type and pile installation procedure. The
Figure 5 P
 erformance of pile capacity methods across depths at ANAM3 method appears suitable for all pile and
soil types.
350
qt = kbqc(avg)(3)
300

250 The Tumay and Fakhroo (1982) method

Pile capacity (kN)

Tumay and Fakhroo (1982) proposed a

200
method for predicting the ultimate pile
capacity of piles in clay soils. Their method
150
is similar to the Schmertmann method,
100 and the unit tip bearing capacity qt is
determined from Equation 4 where qc1 is
50 the average of qc values 4D below the pile
tip, qc2 is the average of the minimum qc
0
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
values 4D below the pile tip, and qa is the
Depth (m) average of the minimum of qc values 8D
Schmertmann De Ruiter LCPC Tumay & Fakhroo above the pile tip. They suggested an upper
Price & Wardle Philipponnat Penpile Aoki & De Alencar limit of 15 MPa for the unit pile tip bearing
capacity qt. Hence, in using this method,
Figure 6 P
 erformance of pile capacity methods across depths at DELTA the pile capacity cannot exceed 15 MPa.

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 49
 depending on whether the soil is gravel,
300
sand, silt or clay. The cone tip resistance
qca is determined from Equation 6.
250
qca(A) + qcb(B)
qca = (6)
200
Pile capacity (kN)

150 Where: qca(A) is the average cone tip resist-

ance within a distance 3D above the pile tip
100 and qcb(B) is the average cone tip resistance
within a distance 3D below the pile tip.
50 The method is applicable to gravels, sands,
silts and clays.
0
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 The Penpile method
Depth (m) The Penpile method (Clisby et al 1978) is
Schmertmann De Ruiter LCPC Tumay & Fakhroo
applicable to sands and clays, and the unit
Price & Wardle Philipponnat Penpile Aoki & De Alencar
tip bearing capacity of the pile

⎧ 0.25qca(tip)f or pile tip in clay

qc = ⎨
Figure 7 P
 erformance of pile capacity methods across depths at IMO1

⎩ 0.125qca(tip)f or pile tip in sand

300
Where: qc is the average of three cone tip
250 resistances close to the pile tip.
A tabular presentation of all the direct
200
methods of estimating pile capacity using
Pile capacity (kN)

the CPT logs is as shown in Table 3 (p 47).

150
Data analysis
A statistical package for social science
100
(SPSS 16.0) was used to analyse the data
generated. The data, comprising pile capac-
50
ity values, pile capacity methods, depth of
pile, and location, was coded into the SPSS
0
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
software. Statistical parameters like mean,
Depth (m) standard deviation, univariate analysis of
Schmertmann De Ruiter LCPC Tumay & Fakhroo variance, and post-hoc tests such as least
Price & Wardle Philipponnat Penpile Aoki & De Alencar square differnce (LSD) and Bonferroni
comparisons were executed.
Figure 8 P
 erformance of pile capacity methods across depths at IMO2
Statistical parameters for choosing
qc1 + qc2 qa is also 15 MPa. The method can only be the most suitable method
qt = + (4)
4 2 applied to sands, silts and clays. The statistical parameters of interest which
will aid in determining which of the pile
The Aoki and De Alencar (1975) method The Price and Wardle (1982) method capacity methods is most suited for the
The method of determining the ultimate The relationship qt = K bqc was established area of interest are the general linear model
load carrying capacity of a pile from CPT by Price and Wardle (1982) as a means of univariate analysis of variance and the
data as presented by Aoki and De Alencar determining the ultimate load-carrying post-hoc tests.
(1975) is given by Equation 5. The unit tip capacity of a pile from CPT data. The The general linear model (GLM) is a
bearing capacity qt is obtained as follows: factor K b is equal to 0.53 for driven piles model which adjusts for interactions of
and 0.62 for jacked piles. Similarly, an covariates with the given factors, while
qca(tip) upper limit of 15 MPa is also imposed on the univariate GLM is the version of the
qt = (5)
Fb the determined value of the ultimate end- general linear model used to implement
bearing pile capacity. The method is thus the analysis of variance (ANOVA) test. The
Where: qca(tip) is the average cone tip primarily applicable to jacked, bored and process deals with a situation where there
resistance around the pile tip, and Fb is an driven piles in stiff (London) clay. is one dependent variable and one or more
empirical factor which depends on the pile independent variables. The general linear
type. Fb ranges from 3.5 to 1.75, depend- The Philipponnat (1980) method model (GLM) implementation of ANOVA
ing on whether the pile is a bored, Franki, Philipponnat (1980) established that supports parameters such as “main effect”
steel or precast pile. The upper limit for qc qt = K bqca where K b varies from 0.35–0.5 and “interaction effect”. The main effect

50 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
is the direct effect of an independent vari-
1 000
able (depth, location of pile and method of
analysis) on the dependent variable (pile 900
capacity), while the interaction effect is the 800

Estimated marginal means

joint effect of two or more independent
700
variables on the dependent variable. The key
statistic in ANOVA is the F-test which tests 600
for difference in group means, i.e. testing if 500
the means of the groups formed by values of
400
the independent variable or combinations of
values for multiple independent variables are 300
significant. If the group means do not differ 200
significantly it implies that the independ-
100
ent variable did not have an effect on the
0
dependent variable. But if the F-test shows IMO3 ANAM1 ANAM2 ANAM4
that the independent variable is related to Location
the dependent variable, then post-hoc tests Schmertmann De Ruiter LCPC Tumay & Fakhroo
are used for further examination. Price & Wardle Philipponnat Penpile Aoki & De Alencar
Post-hoc analyses are usually concerned
with finding patterns and or relationships Figure 9 C
 omparative performance of pile capacity methods in cohesionless soils across all
between subgroups of sampled populations depths and locations
that would otherwise remain undetected
and undiscovered using earlier statistical
450
methods (Wikipedia 2014). The tests are
designed for situations where a significant 400
omnibus F-test with a factor that consists
350
of three or more means has been obtained,
Estimated marginal means

and additional exploration of differences 300

among means is needed to provide specific
information on which means are signifi-
cantly different from one another. Two
popular post-hoc tests that will be consid-
ered in this work are the least significant
difference (LSD) test and the Bonferroni
test. The LSD test explores all pair-wise
comparisons of means comprising a factor
using the equivalent of multiple T-tests, i.e.
it determines what the smallest difference
between means would be for the compari-
son to be statistically significant. A limita-
tion of the LSD is that it makes no adjust-
ment for the fact that multiple comparisons
are being made. The Bonferroni test, on
the other hand, multiplies each of the
significant levels from the LSD test by the
number of tests carried out and, as such,
mean differences that were significant in
LSD could become non-significant after
the completion of the Bonferroni tests.
RESULTS AND DISCUSSIONS
Comparative performance of the pile
capacity determination methods
The figurative illustration of the com-
parative performance of the various pile
capacity determination methods across
the various depths and locations is shown
in Figures 1 to 8, while the comparative
250
200
150
100
50
0
IMO1
Schmertmann
Price & Wardle
Figure 10 C
 omparative performance of pile capacity methods in cohesive soils across all depths
and locations
summary of the pile capacity determina-
tion methods across locations is shown
in Figures 9 and 10. In these figures the
estimated marginal means of the pile
capacities determined by the various meth-
ods are used as a basis of comparison. The
estimated marginal means are basically the
mean of a group or subgroup’s measures
of a variable in an experiment, and is
important when comparing the means of
unequal sample sizes where one takes into
consideration each mean in proportion to
its sample size. Thus, by observation, one
can see that for cohesionless soils, the Aoki
and De-Alencar, Philipponnat, LCPC, and
Price and Wardle methods had central
IMO2 Delta ANAM3 8.0
Location
De Ruiter LCPC Tumay & Fakhroo
Philipponnat Penpile Aoki & De Alencar
values which are also close (Figure 9).
The other methods (Schmertmann and
De Ruiter, Penpile, and the Tumay/Fakhroo
methods) had either very high or very low
values of pile capacity. For cohesive soils
it is the LCPC, Tumay and Fakhroo, De
Ruiter and Penpile methods that are in the
central range (Figure 10). The others had
either very high values implying lower rein-
forcements for the piles, or very low values
implying very high reinforcements. The
generally perceived good performance of
the LCPC, Tumay and Fakhroo, De Ruiter
and Penpile methods in clays is believed
to be as a result of the fact that the LCPC
considered almost all types of piles and

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 51
Table 4 UNIANOVA test between subject effects for cohesive soils
Dependent variable: Pile capacity Significance
Degree of freedom Mean square F-statistic
Source Type III sum of squares (0.05)

Model 1.022E7a 104 98266.407 179.385 .000

Depth 45785.494 9 5087.277 9.287 .000

Location 445866.984 3 148622.328 271.309 .000

Treatment 1408059.670 7 201151.381 367.200 .000

Location* treatment 140814.543 21 6705.454 12.241 .000

Depth* treatment 38886.600 63 617.248 1.127 .264

Error 118324.190 216 547.797

Total 1.034E7 320

a: R squared = .989 (adjusted R squared = .983)

*: Versus

soil conditions in its derivation (197 pile
load tests were carried out), while the
Tumay and Fakhroo method had clays as
its primary target. The De Ruiter method
considered the popular pile-bearing capac-
ity factor Nc with qt determined from
average values of cone resistances around
the pile tip, while the Penpile methods
considered the average of three cone tip
resistances around the pile tip. Similarly,
the Philipponnat method, just like the
LCPC method, considered a wide range of
soil types, and that could be the reason for
its good performance.
Univanova test between
subject effects (for cohesive
and cohesionless soils)
The UNIANOVA tests show that a sig-
nificant omnibus F-test with a factor that
consists of three or more means has been
obtained. The adjusted R-squared values
for the models used are 0.983 and 0.987 for
cohesive and cohesionless soils respectively,
as can be seen in Tables 4 and 5. It can also
be seen from these tables that the interac-
tions between location and treatment
(methods of pile capacity prediction), depth
and treatment, and location and depth
were statistically significant for cohesion-
less soils, but the interaction between
depth and treatment was not statistically
significant for cohesive soils. This implies
that, for the cohesive soils under investi-
gation, the effect of treatment on depth
was not consistent (level of significance
for depth vs treatment is 0.264), and this
could be attributed to the complexities of
the clay minerals, as all the layers were of
varying property.
Post-hoc tests
The ANOVA test results shown in Tables 4
and 5 only show that the interactions
between treatment/location, depth/treat-
ment and depth/location are significant. It
does not compare the interactions between
the various treatments, thus the essence of
the post-hoc tests. The LSD and Bonferroni
tests were carried out for the two groups of
soils and the results shown in Tables 6 to
9, with some of the results for the methods
with significant differences omitted. The
omissions were made because the exhaus-
tive tables are only a repetition of what is
presented, as is the case with Tables 6 and
8 where only the mean differences between
the De Ruiter and Penpile methods are
not significant. From Tables 6 and 7 one
can see that both the LSD and Bonferroni
tests agree for cohesive soils. The results

Table 5 UNIANOVA test between subject effects for cohesionless soils

Dependent variable: Pile capacity
Degree of freedom Mean square F-statistic Significance (0.05)
Source Type III sum of squares

Model 5.920E7a 131 451877.577 179.637 .000

Depth 1605171.150 9 178352.350 70.901 .000

Location 1666758.884 3 555586.295 220.864 .000

Treatment 1.699E7 7 2427403.560 964.975 .000

Depth* treatment 1176787.123 63 18679.161 7.426 .000

Location* treatment 1209323.771 21 57586.846 22.893 .000

Depth* location 224085.719 27 8299.471 3.299 .000

Error 475431.182 189 2515.509

Total 5.967E7 320

a: R squared = .992 (adjusted R squared = .987)

*: Versus

52 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
Table 6 Post-hoc tests (LSD) for pile capacity methods in cohesive soils
Multiple comparisons

Pile capacity LSD Mean difference 95% Confidence interval

Standard error Significance
(I) Treatment (J) Treatment (I-J) Lower-bound Upper-bound

Schmertmann –148.75* 3.899 .000 –156.44 –141.06

LCPC –52.08* 3.899 .000 –59.77 –44.39

Price and Wardle –132.09* 3.899 .000 –139.78 –124.39

De Ruiter Philipponnat 27.31* 3.899 .000 19.62 35.00

Penpile –4.60 3.899 .239 –12.29 3.09

Aoki and De-Alencar 46.03* 3.899 .000 38.34 53.72

Tumay and Fakhroo –37.25* 3.899 .000 –44.95 –29.56

Schmertmann –144.15* 3.899 .000 –151.84 –136.46

De Ruiter 4.60 3.899 .239 –3.09 12.29

LCPC –47.48* 3.899 .000 –55.17 –39.79

Penpile Price and Wardle –127.48* 3.899 .000 –135.18 –119.79

Philipponnat 31.91* 3.899 .000 24.22 39.60

Aoki and De-Alencar 50.63* 3.899 .000 42.94 58.33

Tumay and Fakhroo –32.65* 3.899 .000 –40.34 –24.96

Based on observed means

The error term is “mean square” (error) = 304.098

* The mean difference is significant at the 0.05 level

obtained from the De Ruiter and Penpile
methods are not significantly different
(note parameters without asterisks under
the mean difference column, whose signifi-
cant values are greater than 0.05). In addi-
tion to that, the Bonferroni test also shows
that there are no significant differences
between the Schmertmann and Price/
Wardle methods, between the LCPC and
Tumay/ Fakhroo methods, and between
the Philipponnat and Aoki/De-alencar
methods. One can therefore say that the
methods of De Ruiter and Penpile, and the
methods of LCPC and Tumay/Fakhroo
may be best suited for tropical cohesive
soils. Considering Tables 8 and 9, it is
also evident that the LSD and Bonferroni
tests agree that there are no significant
differences between the methods of
Schmertmann and De Ruiter. In addition,
Bonferroni’s method also shows that there
are no differences between the LCPC and
Philipponnat methods. It is therefore the
authors’ opinion that, considering safety
and economy, the LCPC and Philipponnat
methods performed better for tropical
cohesionless soils.

Table 7 Post-hoc tests (Bonferroni) for pile capacity methods in cohesive soils
Multiple comparisons

Pile capacity Bonferroni Mean difference 95% Confidence interval

Standard error Significance
(I) Treatment (J) Treatment (I-J) Lower-bound Upper-bound

Schmertmann Price and Wardle 16.66 7.048 .523 –5.55 38.88

De Ruiter Penpile –4.60 7.048 1.000 –26.82 17.61

LCPC Tumay and Fakhroo 14.83 7.048 1.000 –7.39 37.04

Price and Wardle Schmertmann –16.66 7.048 .523 –38.88 5.55

Philipponnat Aoki and De-Alencar 18.72 7.048 .233 –3.49 40.94

Penpile De Ruiter 4.60 7.048 1.000 –17.61 26.82

Aoki and De-Alencar Philipponnat –18.72 7.048 .233 –40.94 3.49

Tumay and Fakhroo LCPC –14.83 7.048 1.000 –37.04 7.39

Based on observed means

The error term is “mean square” (error) = 993.418

* The mean difference is significant at the 0.05 level

Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018 53
Table 8 Post-hoc tests (LSD) for pile capacity methods in cohesionless soils

Multiple comparisons

Pile capacity LSD 95% Confidence interval

Mean difference
Standard error Significance
(I-J)
(I) Treatment (J) Treatment Lower-bound Upper-bound

De Ruiter .0000 11.21497 1.000 –22.1226 22.1226

LCPC 394.2745* 11.21497 .000 372.1519 416.3971

Tumay and Fakhroo 682.2902* 11.21497 .000 660.1677 704.4128

Schmertmann Price and Wardle 454.4270* 11.21497 .000 432.3044 476.5496

Philipponnat 367.9948* 11.21497 .000 345.8722 390.1173

Penpile 578.4968* 11.21497 .000 556.3742 600.6193

Aoki and De-Alencar 285.5832* 11.21497 .000 263.4607 307.7058

Schmertmann .0000 11.21497 1.000 –22.1226 22.1226

LCPC 394.2745* 11.21497 .000 372.1519 416.3971

Tumay and Fakhroo 682.2902* 11.21497 .000 660.1677 704.4128

De Ruiter Price and Wardle 454.4270* 11.21497 .000 432.3044 476.5496

Philipponnat 367.9948* 11.21497 .000 345.8722 390.1173

Penpile 578.4968* 11.21497 .000 556.3742 600.6193

Aoki and De-Alencar 285.5832* 11.21497 .000 263.4607 307.7058

Based on observed means

The error term is “mean square” (error) = 2515.509

* The mean difference is significant at the .05 level

Table 9 Post-hoc tests (Bonferroni) for various CPT pile prediction methods in cohesionless soils
Multiple comparisons

Pile capacity Bonferroni Mean difference 95% Confidence interval

Standard error Significance
(I) Treatment (J) Treatment (I-J) Lower-bound Upper-bound

Schmertmann De Ruiter .0000 11.21497 1.000 –35.5376 35.5376

De Ruiter Schmertmann .0000 11.21497 1.000 –35.5376 35.5376

LCPC Philipponnat –26.2797 11.21497 .564 –61.8174 9.2579

Schmertmann –367.9948* 11.21497 .000 –403.5324 –332.4571

Philipponnat De Ruiter –367.9948* 11.21497 .000 –403.5324 –332.4571

LCPC 26.2797 11.21497 .564 –9.2579 61.8174

Based on observed means

The error term is “mean square” (error) = 2515.509

* The mean difference is significant at the .05 level

CONCLUSION whose values are very high. It can also REFERENCES

It is therefore possible to conclude from the
preceding discussions, and from Figures 9
and 10, that the LCPC and Philipponnat
methods are best suited for cohesion-
less soils, as they are the two methods
whose predicted pile capacity values are
consistent and average, compared with
the Schmertmann and De Ruiter methods
be concluded that the LCPC, Tumay/
Fakhroo and De Ruiter methods gave the
best results and are thus recommended
for cohesive soils within the study area.
Specifically, the LCPC method is, interest-
ingly, suitable for both classes of soils and
this agrees with the studies on Louisiana
soils by Titi et al (1999).
Abu-Farsakh, M Y & Titi, H H2004. Assessment of
direct cone penetration test methods for predicting
the ultimate capacity of friction driven piles. Journal
of Geotechnical and Geoenvironmental Engineering,
130(9): 935–944.
Adeniji, F A 1991. Recharge function of vertisolic
vadose zone in sub-Sahelian Chad Basin.
Proceedings, 1st International Conference on Arid

54 Volume 60 Number 1 March 2018 Journal of the South African Institution of Civil Engineering
 Zone Hydrology and Water Resources, Maiduguri,
Nigeria, pp 331–348.
Andriesse, J P 1988. Nature and management of
tropical peat soils. FAO Bulletin 59. Rome: FAO –
Food and Agriculture Organization of the United
Nations.
Anon 1990. Tropical residual soils. Report of a working
party of the Engineering Group of the Geological
Society. Quarterly Journal of Engineering Geology,
23: 1–101.
Aoki, N & De Alencar, D 1975. An approximate
method to estimate the bearing capacity of piles.
Proceedings, 5th Pan-American Conference of Soil
Mechanics and Foundation Engineering, Buenos
Aires, Vol. 1, pp 367–376.
Bustamante, M & Gianeselli, L 1982. Pile bearing
capacity predictions by means of static
penetrometer CPT. Proceedings, 2nd European
Symposium on Penetration Testing (ESOPT-II),
Amsterdam, Vol. 2, pp 493–500.
Chen, F H 1975. Foundations on Expansive Soils.
Amsterdam: Elsevier Scientific.
Clisby, M B, Scholtes, R M, Corey, M W, Cole, H A,
Teng, P & Webb, J D 1978. An Evaluation of Pile
Bearing Capacities, Vol. I. Final Report. Mississippi
State Highway Department.
Culshaw, M G, Hobbs, P R N & Northmore, K J 1992.
Engineering geology of red tropical clay soils: Manual
sampling methods. Technical Report WN/93/14.
Nottingham, UK: British Geological Survey.
De Ruiter, J & Beringen, F L 1979. Pile foundations for
large North Sea structures. Marine Geotechnology,
3(3): 267–314.
Eslami, A & Gholami, M 2006. Analytical model for
the ultimate bearing capacity of foundations from
cone resistance. Scientia Iranica, 13(3): 223–233.
Eslami, A & Fellenius, B H 1995. Toe bearing capacity
of piles from cone penetration test (CPT) data. Paper
presented at the International Symposium on Cone
Penetrometer Testing (CPT ’95), Linköping, Sweden,
4–5 October 1995.
Gogo-abite, I A 2005. Slope stability analysis of
laterite soil embankments. Unpublished Master’s
dissertation. Orlando, FL: University of Central
Florida, Department of Civil and Environmental
Engineering.
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
Hobbs, P R N, Entwisle, D C, Northmore, K J &
Culshaw, M G 1992. Engineering geology of red
tropical clay soils. Geotechnical characterisation:
Mechanical properties and testing procedures.
Technical Report WN/93/13. Nottingham, UK:
British Geological Survey.
ITA (Information Technology Associates) USA 2011.
Laterites. Available at: http://www.theodora.com/
encyclopedia/l/laterite.html (accessed in December
2014).
Luckham, P F & Rossi, S 1999. The colloidal and
rheological properties of bentonite suspensions.
Advances in Colloid Interface Science, 82: 43–92.
Morin, W J & Todor, P C 1976. Laterite and lateritic
soils and other problem soils of the tropics: An
engineering evaluation and highway design study for
United States, Vol. II. Baltimore, MA: Agency for
International Development, pp 1–2.
Nnadi, G 1987. Geotechnical properties of tropical
residual soils. Research Report. Luleå, Sweden: Luleå
University of Technology.
Nottingham, L C 1975. Use of quasi-static friction
cone penetrometer data to predict load capacity
of displacement piles. PhD thesis. Gainesville,
FL: University of Florida, Department of Civil
Engineering, p 553.
Ogunsanwo, O 1990. Geotechnical properties of
undisturbed and compacted amphibolite derived
laterite soil. International Engineering Geology,
31(2): 183–188.
Philipponnat, G 1980. Methode Pratique de Calcul
d’un Pieu Isole a l’aide du Penetrometre Statique
[Practical method of calculating an isolated pile,
using the static penetrometer]. Revue Francaise de
Geotechnique, 10: 55–64.
Price, G & Wardle, I F 1982. A comparison between
cone penetration test results and the performance
of small diameter instrumented piles in stiff
clay. Proceedings, 2nd European Symposium on
Penetration Testing, Amsterdam, Vol. 2, pp 775–780.
Punke, E 2014. Characteristics of tropical soils.
Available at: http://www.ehow.com/list_6663526_
characteristics-tropical-soils.html (accessed in
December 2014).
Rahardjo, H, Aung, K K, Leong, E C & Rezaur, R B
2004. Characteristics of residual soils in Singapore
as formed by weathering. Engineering Geology, 73:
157–169.
Schmertmann, J H 1953. Estimating the true
consolidation behaviour of clay from laboratory test
results. Proceedings of the ASCE, 79(311): 26.
Schmertmann, J H 1978. Guidelines for cone
penetration test (Performance and design). Report
No. FHWA-TS-78-209. Washington, DC: U.S.
Department of Transportation, p 145.
Shooshpasha, I, Hasanzadeh, A & Taghavi, A 2013.
Prediction of the axial bearing capacity of piles
by SPT-based and numerical design methods.
International Journal of GEOMATE, 4(2): 560–564.
Terzaghi, K & Peck, R B 1967. Soil Mechanics in
Engineering Practice, 2nd ed. New York: Wiley
International.
Titi, H H, Murad, P E & Abu-Farsakh, M Y 1999.
Evaluation of bearing capacity of piles from cone
penetration test data. A report prepared for the
Louisiana Transportation Research Centre, Baton
Rouge, LA, USA. Project No. 736-99-0533. p 115.
Tomlinson, M J & Boorman, R 1999. Foundation
Design and Construction, 6th ed. Harlow, UK:
Longman, p 536.
Townsend, F C, Manke, P G & Parcher, J V 1973. The
influence of sesquioxides on lateritic soil properties.
Highway Research Record, 374: 80–92.
Tumay, M T & Fakhroo, M 1982. Friction pile capacity
prediction in cohesive soils using electric quasi-
static penetration tests. Interim Research Report
No. 1. Baton Rouge, LA: Louisiana Department of
Transportation and Development, Research and
Development Section.
Wesley, L 2009. Behaviour and geotechnical properties
of residual soils and allophane clays. Obras y
Proyectos, 6: 5–10.
Wikipedia 2014. Post-hoc analysis. Available at:
http://en.wikipedia.org/wiki/Post-hoc_analysis
(accessed in December 2014).
Wust, R A J, Bustina, R M & Lavkulich, L M 2003.
New classification systems for tropical organic-rich
deposits based on studies of the Tasek Bera Basin,
Malaysia. Catena, 53: 133–163.
Zhou, Z 1995. Construction and application of clay-
swelling diagrams by use of XRD methods. Journal
of Petroleum Technology, 474: 306.
55
DISCUSSION
Journal of the South African
Institution of Civil Engineering
ISSN 1021-2019
Vol 60 No 1, March 2018, Pages 56–57, Paper 1200
Publishing particulars of
the paper under discussion
Vol 59 (1) 2017, Pages 11–21, Paper 1200:
Planning for desalination in the context
of the Western Cape water supply system
(http://dx.doi.org/10.17159/2309-8775/2017/v59n1a2)

C L Blersch, J A du Plessis

COMMENT Equation 2 by Van Niekerk and Du Plessis

The authors must be congratulated on
demonstrating the introduction of a sea-
water desalination plant as an integral part
of a water supply system – in this case the
Western Cape water supply system. While
this new resource would provide a constant
and reliable source of water, its operating
costs typically are high in comparison
with the cost of other traditional resource
developments. As the authors showed (e.g.
in Figure 7 of their paper), due to the sto-
chastic nature of the hydrology on which
the traditional resources depend, the aug-
mentation requirement from a desalination
plant with its high operating cost will be of
a stochastic nature as well. This follows as
the annual operating decisions, based on
system risk analyses, can be expected to
favour resources with lower operating costs
in times when the system is relatively flush
and the risk of curtailments in the short
term is expected to be within acceptable
limits. The authors therefore demonstrated
that, by simulating the annual operating
decisions according to the methodology
proposed by Van Niekerk and Du Plessis
(quoted reference 2013a), the operating
costs would also vary stochastically over
time. These variable costs then carry over
into the determination of the unit reference
values (URVs), showing the latter them-
selves to be uncertain in nature.
It is with the authors’ calculation of
the URVs that this writer has a difficulty.
While seemingly the improved URV deter-
mination methodology by Van Niekerk and
Du Plessis (referenced incorrectly as 2013a
– it should have been quoted as 2013b)
is accepted by the authors (2017:14), the
equation for determining the URV in their
Equation 1 (2017:15) is not the same as
(quoted ref 2013b:552), which is:
PV of life cycle costs
URV =
PV of quantity of water
incrementally assured
In the authors’ Equation 1 the denomina-
tor is given as “NPV of the water supplied
from the desalination plant (Wn)”, which
is quite different to the “PV of quantity
of water incrementally assured” in the
above equation.
Van Niekerk (2013) demonstrated that
the URV concept should be grounded
in the cost-effectiveness analysis (CEA)
theory. Van Niekerk and Du Plessis sum-
marised it as follows (quoted 2013b:554):
“Examination of the URV measure in
view of its underlying CEA economic
theory revealed that a good measure of
effectiveness would be the incremental
water availability assured by the expansion
of a system and bounded by the projected
demand curve until full capacity is reached.
The PV of the annual quantities of water
thus assured is used as denominator in the
URV equation.”
Analogous to what was advocated by
Van Niekerk and Du Plessis for inter-basin
transfer systems, a conceptual separation
is required between the water supplied by
the desalination plant, affecting operating
costs, and the water incrementally assured,
used as a proxy measure for effectiveness.
The authors did not make such a sepa-
ration. It is suggested that the URV cal-
culations be redone with the denominator
amended in accordance with the meth-
odology of Van Niekerk and Du Plessis
(quoted 2013b). The expectation is that
the URV results will be quite different,

Blersch CL, Du Plessis JA. Planning for desalination in the context of the Western Cape water supply.
56 J. S. Afr. Inst. Civ. Eng. 2018:60(1), Art. #1200, 2 pages. http://dx.doi.org/10.17159/2309-8775/2018/v60n1a6
be revelatory regarding the optimality of
desalination plants in terms of sizing and
operations, and give rise to some differ-
ent conclusions from those arrived at in
the paper.
The difference in URV calculations
can be illustrated by taking the example
of the 150 Mℓ/d desalination plant with
the 70% storage level trigger (scenario D
(70%)). The incremental 1:50 year yield of
this plant is about 120 Mℓ/d (44.106 m3/a)
(estimated from Figure 5), providing about
two to three years of augmented supply
before a follow-up intervention is required
(deduced from Table 3), and the median
(50 percentile) supply over the 20 years
(from Figure 6) is estimated as 72 Mℓ/d
on average. The denominator in this case
should be the incrementally assured supply
over the economic life of the interven-
tion (assumed here to be 20 years to fit in
roughly with the planning horizon). While
the discounted volume of water, used as
denominator to determine the URV of
R15.49/m3 in Figure 4, was not stated in
the paper, logically it had to relate to the
72 Mℓ/d figure, whereas it should have
been associated with the 120 Mℓ/d, being
the approximate incremental assurance
provided by the desalination plant, in keep-
ing with the approach by Van Niekerk and
Du Plessis. The correct median URV for
the 150 Mℓ/d plant, operated at a 70% trig-
ger level, would be closer to an estimated
R9.30/m3 than the quoted R15.49/m3.
A further note for correction: The title
at the top, within the border, of Figure 5,
referring to historic firm yield, has presum-
ably been incorrectly copied from Figure 4.
It should refer to the 1 in 50 year yield,
as in the title given below the border for
Figure 5.
Dr Peter van Niekerk Pr Eng FSAICE
peter.v.n@mweb.co.za
References
Blersch, C L & Du Plessis, J A 2017. Planning for
desalination in the context of the Western Cape
water supply system. Journal of the South African
Institution of Civil Engineering, 59(1): 11–21.
Van Niekerk, P H 2013. Hydrologic-economic appraisal
of inter-basin water transfer projects. PhD thesis,
Stellenbosch University. URL: http://hdl.handle.
net/10019.1/79887.
RESPONSE FROM AUTHORS
Dr Peter van Niekerk has raised a distinc-
tion between the denominator used in the
Journal of the South African Institution of Civil Engineering Volume 60 Number 1 March 2018
equation for calculation of Unit Reference
Values (URVs) in the paper (Equation 1)
and that used in the work of Van Niekerk
and Du Plessis which was the reference
point for this equation. Dr Van Niekerk
explained that the value that should
be used according to his methodology
is associated with the volume of water
incrementally assured. With reference to
our paper, this value would be the present
value of the incremental water availability
assured by the expansion of the system,
being the 1:50 year yield calculated using
the equations shown on Figure 5 or using
Equation 3, bounded by the projected
demand curve depicted in Figure 2. On the
other hand, the value used in the denomi-
nator of the URV equation to calculate
URVs (as shown in Table 4 of our paper) is
the present value of the annual volume of
water supplied over the 20-year horizon.
For example, the 50th percentile URV
values in Table 4 were calculated using the
present value of the 50th percentile supply
values over the 20-year period as presented
in Figure 6.
On reviewing Dr Van Niekerk’s expla-
nation, he is correct that there is indeed a
distinction between the values used in the
two papers (and that they might yield dif-
ferent results). However, we do not believe
that the approach adopted in our paper is
incorrect, and is still of value. Updating
the results using the volume of water
incrementally assured for comparison
could be the subject of further research.
For now, we propose that the third and
fourth sentences of the second paragraph
under “Cost estimation and calculation
of URVs” on page 14 be amended to
read “The URV approach, as presented
by Hoffman and Du Plessis (2008), was
adapted by Van Niekerk and Du Plessis
(2013a) by using the actual volume of
water supplied based on a stochastic analy-
sis in the WRPM to calculate the operating
costs in the nominator, and the volume of
water incrementally assured by the project
in the denominator, to calculate the URVs
of inter-basin transfers. This approach
was adapted for application in this paper
by, instead, using the modelled annual
volumes of water supplied from the desali-
nation plant, as extracted from the WRPM
analyses, to calculate costs.”
In addition, with reference to Equation 1,
it was in fact the Present Value (PV) of costs
that was used in the calculations, not the
Net Present Value (NPV) as was stated in
the paper.
Lastly, the correction regarding Figure 5
is noted. This title should read “Increase in
1 in 50 year yield”.
Catherine Blersch
catherine.blersch@aurecongroup.com
Prof Kobus du Plessis
jadup@sun.ac.za
DR VAN NIEKERK’S COMMENT
TO AUTHORS’ RESPONSE
The URV approach of Van Niekerk and Du
Plessis was solidly grounded in the eco-
nomic theory of cost-effectiveness analysis.
The departure by Blersch and Du Plessis, as
is suggested in their proposed amendment,
is neither motivated, substantiated, nor
grounded in economic theory.
Both approaches cannot be correct. An
erroneous application of the URV meth-
odology leads to fundamentally flawed
results. Blersch and Du Plessis’s conclusions
are based on such results. I consider their
paper as important for the insights that it
can bring in considering the inclusion of a
desalination plant in a water resource sys-
tem. I would therefore urge the authors to
urgently apply the necessary corrections.
AUTHORS’ FURTHER RESPONSE
While we agree with the URV approach
suggested in Van Niekerk and Du Plessis, we
do not agree that the deviation as used in the
authors’ (Blersch and Du Plessis) publication
resulted in fundamentally flawed results
based on the objectives of the paper. As stat-
ed in our original response, we agree that it
will be interesting (and needed) to repeat the
exercise with the approach (with reference
to the using of the volume of water incre-
mentally assured), but in this case study we
raised a number of limitations which might
have impacted on the final results presented
in the paper. It was, however, the stochastic
nature of the volume of water that was sup-
plied (or assured) that was of importance to
illustrate, as was done in the paper.
It is, however, doubtful that the main
findings (that the unit reference values for the
desalination plant decrease as the percentage
supply from the plant increases, and that
the unit reference values decrease with an
increase in desalination plant capacity) will
change, given that the volume of water incre-
mentally assured, based on our analysis, varies
with desalination plant trigger level in a simi-
lar way to the actual volume of water supplied.
57
Guidelines for the preparation of papers and technical notes
Authors should comply with the following guidelines when preparing papers for publication in the journal
The Journal of the South African Institution of Civil
Engineering is published quarterly in March, June,
September and December. Articles submitted for
publication are reviewed by a panel of referees under
the guidance of the SAICE Journal Editorial Panel.
The journal publishes research papers covering all the
disciplines of civil engineering (structural, geotechnical,
railway, coastal/marine, water, construction, environ-
mental, municipal, transportation) and associated topics
that are relevant to the civil engineering profession, and
that preferably have particular relevance to civil engi-
neering in southern Africa and the African continent.
When preparing articles for publication, authors
should please take note of the following and comply
with the guidelines as set out:
CLASSIFICATION OF ARTICLES
CONSIDERED FOR PUBLICATION
■■ Technical papers are well-researched, in-depth,
fully referenced technical articles not exceeding
6 000 words in length (excluding tables, illustra-
tions and the list of references). Related papers that
deal with ‘softer sciences’ (e.g. education, social
upliftment, etc) are accepted if they are of a techni-
cal nature and of particular interest to the civil
engineering profession. The latter type of paper will
be subject not only to peer-review by civil engineers,
but also to review by non-engineering specialists in
the field covered by the paper.
■■ Technical notes are short, fully referenced technical
articles that do not exceed 2 000 words. A typical
technical note will have limited scope often dealing
with a single technical issue of particular importance
to civil engineering.
■■ Review papers are considered for publication as
either technical papers or technical notes on condi-
tion that they are the original work of the author
and will assist the reader with the understanding,
interpreting or applying of the subject under review.
A review paper must contain criteria by which the
work under review was evaluated, and contribute
by synthesising the information and drawing new
conclusions from the dissemination of the previously
published work.
■■ Discussion on published articles is welcomed up to
six months after publication. The length of discus-
sion contributions is limited to 1 500 words. Where
appropriate, discussion contributions will be subject
to the normal reviewing process and will be for-
warded to the authors of the original article for reply.
POLICY REGARDING LANGUAGE AND
ORIGINALITY OF SUBMITTED ARTICLES
■■ Language: Manuscripts should preferably be
presented in English, as the journal is distributed
internationally. Articles submitted in any of the
other official South African languages should be
accompanied by an expanded abstract in English.
■■ Original work: Papers and technical notes must
be original contributions. Authors must affirm
that submitted material has not been published
previously, is not under consideration for publica-
tion elsewhere and will not be submitted elsewhere
while under consideration by the SAICE Journal
Editorial Panel. It is the responsibility of the authors
to ensure that publication of any paper in the journal
will not constitute a breach of any agreement or
the transgression of any law. The corresponding
author should confirm that all co-authors have
read and approved the manuscript and accept these
conditions. Authors are responsible for obtaining
permission to publish experimental data and other
information that may be confidential or sensitive.
Authors are also responsible for obtaining permis-
sion from copyright owners when reproducing mate-
rial that had been published elsewhere. Proof of such
permission must be supplied.
SUBMISSION PROCEDURES
AND REQUIRED FORMAT
■■ Online submission: Manuscripts must be uploaded
as PDF files (http://journal.saice.org.za). Individual
file sizes may not exceed 4 MB. Should you experi-
ence problems uploading your paper, please contact
the editor (verelene@saice.org.za).
■■ Format: Manuscripts should be prepared in
MS Word and presented in double line spacing,
single column layout with 25 mm wide margins. Line
numbers must be applied to the whole document.
All pages should bear the authors’ names and be
numbered at the bottom of the page. With the excep-
tion of tables and figures (see below) the document
should be typed in Times New Roman 12 pt font.
Contributions should be accompanied by an abstract
of not more than 200 words.
■■ First page: The first page of the manuscript should
include the title of the paper, the number of words of
the main text (i.e. excluding figures, tables and the
list of references), the initials and surnames of the
authors, professional status (if applicable), SAICE
affiliation (Member, Fellow, Visitor, etc), telephone
numbers (landline and mobile), and e-mail and
postal addresses. The name of the corresponding
author should be underlined. Five keywords should
be suggested.
■■ Figures, tables, photos and illustrations: These
should preferably be submitted in colour, as the
journal is a full-colour publication.
■■ Their positions should be clearly marked in the
text as follows: [Insert Figure 1].
■■ Figures, tables, photos, illustrations and equa-
tions should be numbered consecutively and
should appear in the text directly after they have
been referred to for the first time.
■■ Illustrations must be accompanied by appropriate
captions. Captions for tables should appear above
the table. All other captions should appear below
the illustration (figures, graphs, photos).
■■ Only those figures and photographs essential to
the understanding of the text should be included.
All illustrations should be referred to in the text.
■■ Figures should be produced using c­ omputer
graphics. Hand-drafted figures will not be
­accepted. Lettering on figures should be equiva-
lent to a Times New Roman 9 pt font or slightly
larger (up to 12 pt) if desired. Lettering smaller
than 9 pt is not acceptable.
■■ Tables should be typed in Times New Roman
9 pt font. They should not duplicate information
already given in the text, nor contain material
that would be better presented graphically.
Tabular matter should be as simple as possible,
with brief column headings and a minimum
number of columns.
■■ Mathematical expressions and presentation
of symbols:
■■ Equations should be presented in a clear form
which can easily be read by non-mathematicians.
Each equation should appear on a separate line
and should be numbered consecutively.
■■ Symbols should preferably reflect those used in
Microsoft Word Equation Editor or Mathtype,
or should be typed using the Times New Roman
symbol set.
■■ Variables in equations (x, y, z, etc, as well as
lower case Greek letters) should be presented
in italics. Numbers (digits), upper case Greek
letters, symbols of metric measurement units (m
for metres, s for seconds, etc) and mathematical/
trigonometrical functions (such as sin, cos and
tan) are not written in italics, but in upright type
(Roman). Variables and symbols used in the body
of the text should match the format used in the
equations, i.e. upright or italics, whichever is
applicable.
■■ Metric measurement abbreviations/units should
conform to international usage – the SI system of
units should be used.
■■ Decimal commas may be used, but decimal
points are preferred.
■■ Symbols should preferably be defined in the text,
but if this is not feasible, a list of notations may be
provided for inclusion at the end of the paper.
■■ Headings: Sections and paragraphs should not be
numbered. The following hierarchy of headings
should be followed:
HEADING OF MAIN SECTION
Heading of subsection
Heading of sub-subsection
■■ References: References should follow the Harvard
system. The format of text citations should be as
follows: “Jones (1999) discovered that …” or “recent
results (Brown & Carter 1985; Green et al 1999)
indicated that …”
 References cited in the text should be listed in
alphabetical order at the end of the paper. References
by the same author should be in chronological order.
The following are examples of a journal article, a
book and a conference paper:
■■ Donald, A W 1954. Pore pressure coefficients.
Géotechnique, 4(4):143–147.
■■ Terzaghi, K 1943. Practical soil mechanics. New
York: Wiley.
■■ Smith, A W, Black, C & White, A 1999. Factors
affecting the strength of soils. Proceedings, ASCE
Conference on Shear Strength of Soils, Colorado,
pp 503–532.
Papers published previously in the Journal of the
South African Institution of Civil Engineering should
be cited if applicable.
■■ Footnotes, trade names, acronyms, abbreviations:
These should be avoided. If acronyms are used, they
should be defined when they first appear in the text.
Do not use full stops after abbreviations or acronyms.
■■ Return of amended papers: Papers requiring
amendments will be accepted up to six months after
the referee reports had been sent to authors, after
which the paper will be withdrawn from the system.
FINAL ARTICLE
■■ Copyright: On acceptance of the paper or technical
note, copyright must be transferred by the author/s
to the South African Institution of Civil Engineering
on the form that will be provided by the Institution.
■■ Photos of authors: The final corrected version of
the paper should be accompanied by recent, high-
resolution head and shoulders colour photographs
and a profile not exceeding 100 words for each of
the authors.
■■ Proofs: First proofs of papers will be sent to authors
in PDF format for verification before publication.
No major re-writes will be allowed, only essential
minor corrections.
 Thanks to our Referees!
The SAICE Journal Editorial Panel would like to thank the persons listed below, all of whom served as referees
during 2017. The quality of our journal is not only a reflection of the level of expertise of participating authors,
but certainly also of the high standard set by our referees.

Mr Tony Aimer Mr Wim Hofsink Dr LDV Roodt

Prof Mark Alexander Dr John Holmes Prof Chris Roth
Dr Joseph Anochie-Boateng Dr Emile Horak Ms Xante Roux
Prof Hans Beushausen Prof Chris James Ms Roslynde Ruch
Prof Billy Boshoff Prof Johan Joubert Prof Manu Santhanam
Mr Izak Breytenbach Prof Roman Kinash Mr Andy Shulze-Hulbe
Dr Nicol Chang Mr Eduard Kleyn Mr Ken Schwartz
Mr Andrew Copeland Dr Christo Krogscheepers Dr Quentin Shaw
Prof Ken Craig Prof Kenny Kwok Dr Marion Sinclair
Dr Peter Day Ms Tanya Lane Mr Givarn Singh
Dr Hennie de Clercq Dr Roman Lenner Prof Giovanni Solari
Mr Johan de Koker Mr Carlo Loubser Prof Raphael Steenbergen
Prof Nick Dekker Dr Hilton Macdonald Dr Breda Strasheim
Dr Erik Denneman Dr Charles Macrobert Dr Peter Strauss
Prof Dimitris Diamantidis Prof James Maina Prof Miroslav Sykora
Mr Willem du Preez Mr Adriaan Meintjes Prof Tiku Tanyimboh
Mr Jacques du Toit Prof B Melville Mr Hugh Thompson
Dr Lucas-Jan Ebels Dr Martin Mgangira Dr Christo van As
Dr Wallied H Elsaigh Mr Derek Mostert Dr Elsabé van Aswegen
Mr Karl Eschberger Prof Pilate Moyo Prof Martin van de Ven
Prof Phil Everett Mr Kenny Mudenda Dr Gert van Rooyen
Prof George Fanourakis Prof Hans-Juergen Niemann Prof Gideon van Zijl
Dr Lloyd Fisher-Jeffes Prof Sunday Nwaubani Dr Sudhir Varma
Mr Amanuel Gebremeskel Mr Pieter Onderwater Mr Ben Verhaeghe
Dr OJ Gericke Prof Mike Otieno Dr Celeste Viljoen
Prof Mitchell Gohnert Dr Phil Paige-Green Dr Eduard Vorster
Prof Chris Guerts Dr George Papageorgiou Mr Kurt Waelbers
Prof Junke Guo Mr Bryan Perrie Dr Richard Walls
Dr Trevor Haas Dr Paul Roberts Prof Mohamed Ya