See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265452431

Performance-Based Contracts and Their Impact on Construction and

Maintenance Practices: A Contractor's Perspective

Article

CITATIONS READS

0 61

4 authors, including:

Bryan Pidwerbesky
Fulton Hogan Ltd
27 PUBLICATIONS   93 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Bryan Pidwerbesky on 15 November 2019.

The user has requested enhancement of the downloaded file.

PERFORMANCE-BASED CONTRACTS AND
THEIR IMPACT ON CONSTRUCTION AND
MAINTENANCE PRACTICES: A
CONTRACTOR’S PERSPECTIVE
Pidwerbesky, B. D.
Fulton Hogan Ltd, Christchurch, New Zealand

SYNOPSIS
Performance contracts are being progressively introduced in New Ze aland to provide contractors with an
opportunity to be innovative and to take greater responsibility (and risk) for both road construction and
maintenance. This paper describes a contractor’s experience with performance contracts, and summarises the
benefits and risks involved.

New Zealand’s performance -based specification for design and construction of flexible pavements was
introduced to foster the use of marginal and non -conforming materials that give similar performance to standard
materials. Another o bjective is to preserve quality aggregates, so that road construction is achieved in a
sustainable manner that mitigates any adverse environmental effects. Two construction projects completed
under these performance specifications are briefly described as case studies.

For the contractor, performance maintenance contracts have instigated a paradigm shift from traditional
production oriented contracting to asset management. The opportunity to drive the decision making process on
treatment investment is both challenging and satisfying. The challenge comes in analysing data, verifying it on
the road and then determining the optimum investment strategy, that of course, is constrained by the budget or
contractor’s annual income.

The paper concludes with example s of how performance contracts encourage development and implementation
of improved, efficient and innovative practices and technologies in both construction and maintenance. This
paper intends to provide an insight into the process so that other road auth orities might also adopt performance
specifications for road construction and maintenance, to encourage optimal use of resources and best practice
to achieve the required pavement performance.
INTRODUCTION
The New Zealand road network totals nearly 95,00 0 km in length, of which 55,000 km have all -weather surfaces,
and serves a population of 4 million over an area of 268,675 km². The state highway network (10,750 km in
length) is managed by Transit New Zealand. Since July 1991, all design, supervision, c onstruction and
maintenance work on state highways has been outsourced. More recently, performance contracts for
maintenance and construction of pavements have been introduced, for a number of reasons, some of which are
discussed in this paper.

There is no general consensus as to the definition of or what constitutes a performance specifications. Two
definitions are:

Specifications that describe how the finished product should perform over time. For highways,
performance is typically described in terms o f changes in physical condition of the surface and
its response to load, or in terms of the cumulative traffic required to bring the pavement to a
condition defined as “failure.” (TRB 2002)

A specification that describes how the end product should perfor m over time. (AUSTROADS 2003)

and performance -based specifications are defined as:

Quality assurance specifications that describe the desired levels of fundamental engineering
properties (e.g., resilient modulus, creep properties, and fatigue properties) that are predictors
of performance and appear in primary prediction relationships (i.e., models that can be used to
predict pavement stress, distress, or performance from combinations of predictors that
represent traffic, environmental, roadbed, and struc tural conditions.) (TRB 2002)

AUSTROADS (2003) recommended the following definition for a performance contract:

A contract that defines obligations and outcomes in terms of performance indicators for the product or
service and that is of sufficient durat ion to enable measurement of the contractor's performance over
time.

Regardless of the definition, there are risks for the road authority and the contractor in using performance
specifications, but the risks can be mitigated by:
• Mutual understanding betwe en the road authority and contractors of the elements of the performance
specifications and contracts;
• Mandatory quality assurance requirements; and ,
• Improved techniques for predicting long term behaviour.

The aspects of performance specifications releva nt to pavement construction are discussed in the next section,
followed by the section on performance contracts for maintenance.

PERFORMANCE SPECIFIC ATIONS FOR CONSTRUCT ION

Instead of developing new pavement material and construction specifications for e very possible alternative
material, Transit aims to encourage innovation and to give contractors the opportunity to trial new materials by
introducing performance specifications. These specifications set out the requirements for materials and road
perform ance (e.g. rutting, roughness etc) at the end of the defects liability period, usually 1 to 3 years after the
road has been opened to traffic. The objective is to allow a pavement to be built utilising a wider range of
materials, provided it can be shown that these materials have adequate strength and durability for the design
life.

The aim of developing performance requirements for the structural design and construction of flexible unbound
pavements is to have a more direct relationship to required in -service performance. In an unbound pavement,
the basecourse and subbase materials are required to:

1. Spread the wheel loads to reduce the load on the soft underlying subgrade and/or other weaker
pavement materials;
 2. Not fail in shear (i.e. shoving or rutting) with the application of wheel loads;
3. Have minimal deformation, where most of the deformation occurs in the subgrade;
4. Not deteriorate structurally over the design life;
5. Adequately hold and support the surfacing; and
6. Not be detrimental to the performance of the surfacing (e.g. cracking).

In New Zealand, the typical pavement consists of a sprayed chip seal over unbound granular base and subbase
layers. The thickness design procedure for thin -surfaced, unbound granular flexible pavements is based on the
AUSTROADS Pavement Design Guide (AUSTROADS 1992), supplemented by Transit (2000). Traditionally all
the other requirements listed are satisfied by using an unbound granular aggregate that complies with New
Zealand’s material specification for basecourse aggregat e, M/4 (Transit 1995). This specification is a recipe for
quarries to make a basecourse that has been proven over time to provide adequate performance in the road.

Materials that are less costly to produce than those complying with M/4 can meet the six requirements. New
Zealand’s performance based specifications B/3 (Transit 2000a) and M/22 (Transit 2000b) provide a framework
to allow these materials to be evaluated and used.

The Performance Based Specification for Structural Design and Construction o f Flexible Unbound Pavements
(referred to as B/3) has been developed to evaluate whether a pavement is meeting the six performance
requirements stated above, at the end of its maintenance period; the acceptance criteria are set at a level that
gives confid ence the pavement will give adequate service over its design life. B/3 requires evidence that the
materials used in the pavement have adequate strength and durability for the design life requested by the client
before construction begins. Notes for the E valuation of Unbound Road Base and Sub -base Aggregates
(referred to as M/22 Notes) were developed to provide guidelines for providing that evidence. M/22 Notes are
not mandatory, and other supporting information proving the suitability of a material, such as road tests, may be
adequate to allow their use. B/3 also includes a number of checks at intermediate points in the design and
construction processes to ensure that the pavement is designed and constructed in a manner that is likely to
produce the desi red performance. The trial procedure is illustrated in Figure 1 , and the process followed by the
Contractor and Engineer is detailed in Table 1.

M/22 uses the generally accepted definition of a basecourse as “the pavement material (stabilised or otherwise )
forming the base, defined as the upper 100 - 200 mm of aggregate in a thin surfaced (less than 35mm)
pavement”. The traffic induced stresses are greatest in this top part of the pavement. Therefore the highest
shear strength and durability is specified for the basecourse.

In M/22, in -service shear strength and deformation resistance of proposed aggregates can be proven by:
• Laboratory testing with the Repeat Load Triaxial (RLT) test;
• Full-scale testing at New Zealand’s accelerated pavement loading facil ity (CAPTIF) which is described
elsewhere (Pidwerbesky 1995 );
• Acceptable performance on roads with documented maintenance and loading histories.

TNZ M/22 defines the sub -base as any pavement material below the basecourse and above the subgrade or
subgrade improvement layer. The overlying basecourse reduces the traffic induced shear stresses in the sub -
base significantly and therefore the requirements are less stringent. The specification allows the use of more
than one sub -base material and recognises that the strength requirements for these materials are governed by
their position in the pavement (depth) and strength assumed for design. As with the basecourse, M/22 requires
that the subbase be durable and strong, however it can, unlike the basecourse, be bound.

Using bound layers in the subbase is acceptable as the unbound overlying basecourse material will prevent any
cracks occurring in a sub -base material reflecting through to the surface. Therefore, there are no limits placed
on the tensile strength or shrinkage of the sub -base material. This allows a sub -base to be a bound material
should the design utilise a bound layer (modulus ≥ 2000 MPa) within the pavement.
 NOTIFICATION MEETING - to discuss: USE
- Site data needed ALTERNATIVE
- Intention to use B/3 on a
- Spec modifications required MATERIALS?
future project
- Process involved & expectations

Yes

MATERIAL DEVELOPMENT
- Locate, develop recipes and test materials for suitability as a base or sub -base.

- If needed, obtain assistance from David Alabaster, TNZ HO.

No
CONFIDENT OF
MATERIALS
No Yes
PERFORMANCE
IN T HE ROAD?

RECEIVE TENDER DOCUMENTS

DESIGN USE ENGINEERS
-PLUS ENGINEERS PAVEMENT DESIGN
No Pre-tender meeting to discuss:
PAVEMENT PAVEMENT DESIGN?
x Engineers pavement design expectations;
x pavement design requirements particular to
the project.

Yes

SUBMIT TENDER
x include own design and materials TENDER
to be used with supporting data; ACCEPTED?
x Engineer to send alternative designs to
Transit NZ Head Office to collate a database.
Yes

CONSTRUCT ROAD
- as per Quality Plan

PASSED
B/3 MAIN TENANCE
INSPECTION?

REPAIR ROAD Yes

No

COMPLIES
- Collect retention's.

Figure 1 Procedure for Evaluating Tenders for B/3 Performance

Specification
Table 1 Step by Step Process for TNZ B/3 Performance -Based Contracts

CONTRACTO R ENGINEER (CLIENT)

1. P r i o r t o a n y r o a d i n g p r o j e c t s b e i n g 2. Site investigation:
advertised, the Contractor in liaison with subgrade strength tests
local quarries should source appropriate deflection tests
pavement materials (modified or otherwise) site survey
and test for compliance with TNZ M/22
Notes for the Evaluation of Un bound Road
Base and Sub -base Aggregates . 3. Initial geometric and pavement design to do
cost estimate.

4. Produce tender documents to include:

• drawing showing plan (hor izontal alignment)
• site data as detailed in Schedule A
• design subgrade strength for each road
section
• calculations for material volumes
5. Advertise tender for contract.

6. Receives tender documents and then:

• sources appropriate pavement
materials (as ab ove);
• designs pavement cross sections based
on site data given in tender documents,
and check for compliance with
Austroads (1992) and NZ Supplement
(2000);
• designs first seal coat;
• determines total costs;
• completes other tender requirements.
7. Submit tender which includes amongst 8. Evaluate tenders including:
other things the following: • checking that the pavement design cross -
• pavement design cross -sections; sections meet or exceed the design life
• pavement materials to be used (include when evaluated using AUSTROAD S (1992)
info needed to identify the material procedures;
types, eg grading) with evidence of • check the pavement materials to be used
compliance with M/22 Notes; comply with M/22 Notes;
• a Quality Plan for constr uction which • chooses and notifies the Contractor;
includes targets for compliance with
appropriate tests.
• unit rates and total cost needed for
payment.
9. Proceeds on site and undertakes site 10. Checks any changed pavement design meets
investigation. or exceeds the design life when evaluated
• obtains agreement with Engineer if site using the procedures in AUSTROADS (1992).
data has changed and then re -designs
pavement cross -section.
11. Proceeds to Construct the pavement to 12. At appropriate times the following items are
their Quality Plan. checked:
• undertakes construction and During construction:
maintena nce to meet the requirements • pavement materials used are correct;
of the specification; • pavement depths achieved;
• pavement layers are compacted to their
t a r g e t d e n s i t i e s.
 CONTRACTO R ENGINEER (CLIENT)

On finished pavement prior to sealing:

• pavement stiffness are achieved;
• construction tolerances are complied with;
• surface shape and rut depth is acceptable;
• roughness is acceptable;
• saturation level is acceptable.
Soon after first coat seal is applied:
• texture depth, chip retention, surface
waterproofness is acceptable.

At the end of the defects liability period (1 to 3

years):
• surface shape and rut depth is acceptable;
• roughness is acceptable;
• texture depth, chip retention, and surface
waterproofness are acceptable

Discussion
B/3 provides the framework to reduce the risk to the client of using new and innovative materials by focussing on
the performance of the pavement at the end of its maintenance period rather than the materials and processes
that go into a producing that pavement. The performance criteria listed in B/3, if achieved at the end of the
defects liability period, give confidence that the pavement will perform adequately for its design life. In addition,
there are a number of intermedi ate checks which are actually either based on the Contractor’s quality assurance
plan or intermediate steps prior to the final acceptance criteria at the end of the defect liability period (1 to 3
years) – these checks are presented in Step 12 of Table 1.

The New Zealand B/3 specification has characteristics of both performance and performance -based
specifications, as defined earlier in this paper, in that the 1 to 3 year warranty period for rut, ride, surface texture,
etc. fits the performance specificati on definition, whereas the use of pavement stiffness, resilient modulus, CBR
and other material properties fits the definition of performance -based specification.

The major risk to the client and contractor is the inadequacy of currently available techniq ues for predicting
pavement performance, both for the materials prior to construction and for the pavement immediately after
construction. Procedures for predicting field performance, especially for recycled or other marginal materials,
must be improved.

M/22 does not specify the use of any stabilisers, grading requirements or number of crushed faces, for example.
It is the contractor’s responsibility to provide a material that will meet the specified durability and strength
requirements. The materials th at meet the basecourse and subbase requirements will range from the
unconventional to minor changes from traditional specifications such as M/4. Cost, availability and risk will be
the driving factors as to the type of material chosen.

CASE STUDIES
To date three pilot projects using B/3 have been completed. On the first project, a realignment on State
Highway 22 south of Auckland, construction was completed in 2001 and project defect liability period ended in
2002. The second project, the realignment on State Highway 6 in the South Island, was completed in April
2002, and completed its defect liability period (12 months) in April 2003. Construction of the third project, the
Napier Hawkes Bay Expressway on the East Coast of the North Island, was complete d in May 2003, which was
followed by a 12 month defect liability period. The second and third projects are discussed below.
Case Study 1: Glenhope Project
This project involved 10 km of road realignment on State Highway 6 from Glenhope to Kawatiri, loca ted 80 km
south of Nelson in the South Island. The project construction period was 21 months and the project was
officially opened on 19 April 2002. The new alignment includes 2 km of passing lanes and has improved the
speed environment to 75 km/h. The i mproved alignment has eliminated a number of accident prone “black
spots”. 4 km of the project passes through Kahurangi National Park, which required some additional attention to
environmental issues.

Local aggregates are alluvial deposits from the Bulle r River; the Gowan quarry is approximately 10 km south of
the site. Fulton Hogan undertook a significant amount of investigation and laboratory work to establish the
suitability of modified local basecourse products compared to aggregate that complies full y with M/4, including
repeat load triaxial testing.

In 2000, State Highway 6 had an Annual Average Daily Traffic (AADT) count of 1500 vehicles per day (vpd),
containing 12% heavy vehicles. For the required 25 year design life, this produces a design traf fic loading of 1.7
6
x 10 equivalent single axle loads (esa). The most recent traffic count gave an AADT of 2070.

The assumed subgrade strength from earlier geotechnical investigations done at the feasibility stage of the
project was a design subgrade CBR of 8. The modulus of the basecourse aggregate was 400 MPa for
pavement design.

Manufactured properties of the proposed quarry material were tested. The only substantial variation from M/4
specification for basecourse aggregate was the Broken Faces crit eria: a minimum of 60% and 40% in the 37.5
to 19 mm and 19.0 to 4.75 mm fraction sizes, respectively, were the contractor’s nominated criteria for the
basecourse, whereas, normally, in order to comply with M/4, 70% of the particles larger than 4.75mm in
basecourse aggregate must have broken faces. The percentage of broken faces actually achieved are in Table
2. For all other properties specified in the contractor’s quality plan, the alternative material used for the
basecourse met or exceeded the minimum requirements of M/4.

Research conducted at New Zealand’s accelerated pavement test track had confirmed that for state highways
carrying low volumes of traffic, the percentage of broken faces in the basecourse aggregate could be reduced
without significant ly affecting the performance of the pavement (Pidwerbesky 1995). The alternative material had
a CBR of greater than 135 (M/4 requires a minimum CBR of 80%), and a modulus of 450 MPa (for RLT test
conditions of 425 kPa deviator stress and 125 kPa confining pressure).

The crushing processes that would have been necessary to fully comply with M/4 is compared to what was
3
required to achieve the specifically designed aggregate material are in Figure 2 . The saving of 30,000 m of raw
feed had significant econom ic advantage as well as environmental benefits, as that quantity of material is still in
the pit that otherwise would have had to be quarried and wasted (there is no potential market for utilising this
waste within an economical haulage distance of the pit ).

After compaction, the subgrade tests included nuclear density meter every 50 m, dynamic cone penetration
(DCP) every 50 m in each lane, and Benkelman Beam deflection every 40 m in each wheelpath in each lane.
This extensive testing regime was done to ensure that the subgrade condition was suitable for the pavement
design above it; maximum allowable values for the deflection and DCP results were 2 mm and 25 mm/blow,
respectively. If the subgrade was deficient, remedies included additional compaction an d replacement of the
material and re -compaction. As a last resort, a third option was to increase pavement layer thicknesses , but this
was not required.
Table 2 Characteristics of Alternative Materials for Glenhope Pavement

Particle Size Range Broken Faces

(mm) (%)
Target in Actual
Quality Plan Achieved
37.5 – 19.0 60 64
19.0 – 9.5 40 52
9.5 – 4.75 40 43

1. Crushing Details for B/3 AP40 (Actually what happened )

AP40

Symons
Rawfeed Prescreen Jaw Cone 55,000m
3

Crusher

3
70,000m AP65
( Some Blending )
( Necessary ) By-Product [ Only some 20mm ]
3
15,000m [ and smaller ]
[ removed ]

2. Crushing Details for M/4 (What it would have been if we had to make M/4)
AP40 M/4

Symons
Rawfeed Prescreen Jaw Cone 55,000m
3

Crushe r

3
100,000m AP65

By-Product [ All 30mm and ]

3
45,000m [ smaller would ]
[ have to be taken ]
[ out ]

Figure 2 Aggregate Production Process for Case Study 1

Pavement construction was done by conventional equipment and techniques. The specified material was
generally a very good material to lay and productivities were no different to what they would have been laying
standard M/4 -compliant material. The material required 6 passes of the rollers to achieve the compaction
density. The pavement laying conditions specified in the contractor’s quality pla n were all achieved.

The primary acceptance criteria for the completed basecourse, as far as the contractor’s pavement designer was
concerned, was a Benkelman Beam deflection of less than 1 mm and a coefficient of variation of less than 25%
(coefficient o f variation is the standard deviation of a sample of date divided by the mean of that data, and is a
measure of uniformity). The pavement was surfaced with a two coat seal.
Post Construction Pavement Performance
The acceptance criteria for pavement perfo rmance after one year were:

Rutting Maximum rut depth of 10 mm

Texture Depth Minimum 1.0 mm
Skid Resistance GripNumber 0.60 (or SFC 0.60)
Roughness Maximum 70 NAASRA counts/km (IRI 2.7) for a 100m section, and average of 55
NAASRA counts/km (IRI 2.1) ov er project length.

Results to date indicate excellent pavement performance.

The texture depth of the finished seal was measured every 80 m in each wheelpath of each lane and centreline
during the maintenance period (post construction) using the sand circ le test, and was in the range of 1.0 to 2.0
mm, with an average of 1.34 mm. The second set of texture depth measurements were done in April 2003, and
the average texture depth had decreased to 1.26 mm, which was substantially greater than the acceptance
criteria of 1.0 mm.

The first high speed data (HSD) survey done by the highway authority, in February 2003, ten months after the
pavement was surfaced, showed that the average surface texture depth over the length of this project was 1.65
mm. The HSD surv ey uses lasers mounted on vehicles traveling at normal highway speeds, so the HSD results
do not compare directly with the manual sand circle test results. The HSD lasers measure texture continuously
and results are an average of the data every 10 m.

Fulton Hogan owns and operates a D -type GripTester, which was used to test the skid resistance of the entire
length during the maintenance period, and the GripNumber was in the range of 0.7 to 0.8, which is an excellent
friction value. Regular skid resistanc e measurements will be done as part of the normal skid resistance survey
conducted annually by Transit NZ over the entire national highway network.

Roughness values for the sealed pavement were extracted from the normal high speed data survey conducted
by Transit NZ annually. The roughness over the length of the project averaged 54 NAASRA counts/km (IRI 2.1),
ten months after opening to traffic.

The rut depths were measured using a manual method every 50 m, four months and 12 months after
construction. Average rut depths have also been extracted from Transit’s annual high speed data (HSD) survey
conducted ten months after construction. The HSD data used is the maximum rut depth that occurred in a 20 m
section of roadway, whilst the manual data is the ma ximum rut depth at each 50 m chainage . The average rut
depth measured by manual and high speed laser techniques after 12 and 10 months, respectively, were 5.4 mm
and 7.8 mm, respectively.

As with all unbound granular pavements subjected to heavy vehicles, the rate of initial rutting under trafficking
immediately after construction is relatively high, but soon levels off to the secondary rate of rutting. Because the
two sets of measurements were not done at the same time, the difference in the two means (HS D vs manual)
cannot be compared. However, the differences in the standard deviations can be compared, and this shows a
significant difference between the two measurement techniques. Even though manual rut depth measurements
are more costly, this form of me asurement is probably more acceptable as a performance measure.

A primary indicator of pavement stiffness (and, thus, its expected performance) is its deflection under load.
Therefore, an extensive programme of Benkelman Beam deflection testing was carri ed out on the subgrade and
pavement layers; deflections were measured every 40 m in each wheelpath of each lane plus along the
2
centreline. B/3 required that 95% of the values measured in a typical 1000 m lot were to be better than or equal
to the target values. However, this still allows for very poor sections to be accepted. One technique to improve
this is to apply a more rigorous statistical analysis. In addition, this statistical analysis can assist in determining a
suitable testing regime for future projects.

The mean pavement deflection was calculated at each chainage by averaging all transverse positions (that is,
each wheelpath in each lane plus the centreline). The target deflection was 1 mm; 99% of the mean deflections
were below 1 mm, while 94 % were less than 0.8 mm. Standard deviations for the subgrade and pavement
deflections were 0.43 mm and 0.19 mm, respectively. As expected, deflections at the pavement surface were
much more consistent than those for the subgrade, yielding a much smaller standard deviation for each
individual section of road as well as for the entire road, as shown in Table 3.

The number of tests per km required to achieve a statistical level of confidence levels were also determined. The
number of deflection tests in an y 1 km section of the road required to provide a 90% confidence level is 17 per
wheelpath, but on average only 10 deflection tests are needed. For a 95% confidence level the average
maximum number of tests required was 38 per km. Thus, a 90% level of con fidence can be statistically
validated by testing deflections every 100 m in each wheelpath.

The confidence intervals used for determining the number of tests required are one tailed, because no lower limit
is required for allowable deflection. In other words, if sampling is carried out using the number of tests specified,
then there will be a 90% (or 95%) certainty that the results are less than the target.

In proposed revisions to the B/3 specification, the requirement that 95% of the material used is better than the
target values will be removed and a more rigorous approach will be adopted. Statistical analysis techniques are
a valuable tool for both the contractor and road authority in quantifying the condition of the pavement and
verifying compliance with acceptance criteria.

Case Study 2: Hawkes Bay Expressway Project

This project extends from the Hawkes Bay Airport on State Highway 2 for 5.3 km. The major items built
included:
• 5.3 km of two lane sealed road;
• Embankment/culvert/bridge crossing ove r the Ahuriri Estuary;
• Roundabouts at the intersections with Tamatea Drive and Taradale Road;
• 750m of two lane arterial road for the Napier City Council.

Construction began in January 2002 and was completed in May 2003. Based on traffic counts taken in t he
existing section of this road, the AADT should be approximately 23,000 vehicles per day (vpd).

Fulton Hogan’s pavement designers generated alternative pavement designs based on the data provided in the
Geotechnical Report at the time of tender and some additional material testing that the contractor undertook pre -
tender. This alternative offered Transit an economic advantage over the original tender design, provided an
opportunity for the contractor and Transit to gain further experience with B/3, and t ransferred risk associated with
granular pavement layers and chip sealing design and construction to the contractor.

The required acceptance criteria are similar to current TNZ granular pavement specifications, except for the
subbase thicknesses, and the key pavement strength parameter, once material gradings and compaction
densities have been achieved, is the stiffness criteria on individual layers: maximum allowable Benkelman Beam
deflections at the top of the finished subgrade and pavement were initiall y specified to be 1.6 and 0.6 mm,
respectively.

At the time of writing of this paper, no performance data was available.

Discussion
There was some difficulty achieving adequate stiffness in the upper sand -silt subgrade material through cement
modificatio n alone, but granular stabilisation of the subgrade improved stiffnesses, so either granular
stabilisation alone or combined granular and cement stabilisation was used for the remaining areas of sand -silt
subgrade. The significant variability of the subgra de and implications on pavement design are summarized in
Table 3. Due to the extremely poor subgrade conditions in some sections of the road, the maximum allowable
deflection for the finished pavement in these sections was relaxed to 1 mm (from 0.6 mm), bu t only if a
coefficient of variation of 20% or less is achieved, to ensure that the pavement is uniform and roughness will not
occur prematurely under trafficking.
Table 3 Subgrade Variability and Pavement Design for Hawkes Bay Expressway

Upper Base Total

Location

Revision
Lower Subbase Thicknes Pavement

Design
Rel a t i v e t o Subgrade

Code
Subbase Thickness s Thickness
Estuary
(mm) (mm) (mm)
North Granular
6
(6.5 x 10 esa) CBR = 15 140 150 290 A
CBR = 8 50 mm 140 150 340 B

Zone 3 Granular 0 200 200 2 C

CH1250 -2000
Zone 3 Sand silt 200 mm thick 0 200 450 5 H
CH1050 -1250 granular layer
& 2000-2100 mixed 250 mm
deep into
subgrade

South Subgrade Subgrade 120 150 470 1 D

6
(3.9 x 10 esa) CBR=5 Stabilised w i t h
2% cement,
200 mm
Zone 4 75 mm thick 120 150 470 3 E
granular layer
mixed 150 mm
deep into
subgrade
200 mm
Zone 6 100 mm thick 0 200 400 6 G
granular layer
+ 2.0% cement
mixed 230 mm
deep into
subgrade
200 mm
Zones 5 & 7 7 5 m m 0 265 465 4 F
granular
overlay + 2.0%
cement mixed
200 mm deep
into subgrade

Revision Details
1 1/5/02 Revised subsequent to site information that subgrade modified with 2% cement
yielded soaked CBR of 14% and that the unmodified in situ subgrade yielded
soaked CBR of 9%.
2 10/9/02 Due to high quality subgrade, pavement reduced to 200mm of M4 in Zone 3
3 10/9/02 Granular stabilisation approved as technique to improve subgrade in lieu of
cement stabilisation .
4 7/11/02 Superceded change mad e in revision 5 (above). Subbase replaced with M/4 to
further increase confidence of performance.
5 7/11/02 For areas where sand silt exists in subgrade. Overlay subgrade with 200mm of
granular then mixed 250 deep.
6 2/12/02 Supersede revision 4 above. F urther improved stabilised layer by increasing
cement content to 2% and treat to 250 mm
This project has highlighted the deficiencies of relying on nuclear density meters for determining compaction
densities. Due to the highly variable nature of the in si tu subgrade material, in situ densities were regularly
checked using sand displacement method, and both solid densities and maximum dry densities were determined
at a much higher frequency than the norm for New Zealand state highway construction, at the in stigation of the
contractor, in order to mitigate the contractor’s risk.

In order to construct the Hawkes Bay Expressway within the contract price, the contractor took advantage of its
national expertise, conducted trials to confirm local materials, modif ied designs and construction techniques
during construction, and optimised allocation of resources and equipment. However, substantial additional costs
were incurred due to waiting for the outcomes of the trials and additional testing required.

A significant aspect of ensuring the successful implementation of performance specification is the culture change
required of both contractor and client. Key staff, such as engineers and supervisors, must be fully briefed and
trained, so that they use the correct approach and understand their respective roles in performance contracts.

PERFORMANCE MAINTENANCE CONTRACTS

In New Zealand, performance maintenance contracts are lump sum Maintenance Contracts of five or ten years
duration, and exclude capital works. The Transit New Zealand (TNZ) performance maintenance contracts
discussed in this paper are of a two year initial term with annual contract roll -overs for the remainder of the five
year duration.

The five year performance maintenance contract is a unique blen d of traditional physical works amalgamated
with asset management skills. Physical works included in such contracts generally include all of the following:
• Cyclic and routine pavement maintenance
• Resurfacing
• Area wide treatments
• Traffic services
• Vegetation control
• Pavement marking
• Incident and emergency response

As a result of becoming involved in asset management, the role of the contractor changes from that of a
predominantly production oriented philosophy to having an understanding of what the asset req uires, when and
by what means it should be implemented in order to achieve full service delivery.

Specialist skills and expertise have to be developed or acquired, in order to be able to interpret High Speed Data
(HSD) results, interrogate road condition databases, perform deterioration modelling and translate the output
into the physical works necessary to achieve the contract Key Performance Measures (KPM’s) and Indicators
(KPI’s), and thus, preserve or enhance the road asset.

Fulton Hogan Ltd currently manages four performance contracts on the state highway network for Transit NZ.
Two of these are in the South Island (both in Canterbury region), and the other two are in the North Island (one
in Waikato region and one in Wellington). The details are show n in Table 4 below.

In traditional maintenance contracts, analysis of data from road condition databases or current high speed data
surveys, and the development, calibration and refining of deterioration models are operations from which
contractors are no rmally excluded. The most exciting element of a performance maintenance contract is being
involved in the decision making process for when to invest, where and how. Performance contracts have
provided a paradigm shift from traditional production oriented c ontracting to asset management and all the skills
and philosophies associated with it.

The opportunity to drive the decision making process on treatment investment is both challenging and satisfying.
The challenge comes in analysing data, verifying it on the road and then determining the optimum investment
strategy that of course is constrained by the budget or contractor’s annual income. The satisfaction comes from
analysis of the following year’s data and the effectiveness of the previous year’s treatmen ts or investment into
asset condition improvement.
Table 4 Performance Contracts

Location Length Description Date

Commenced
Christchurch - 110km Urban Oct 1999
Waimakariri
(Canterbury, South
Island)
Selwyn – Banks 270km Rural Oct 2001
Peninsula
(Canterbury, South
Island)
Matamata – Hauraki 330km Rural Oct 2001
(Waikato, North Island)
Wellington (North Island) 240km Urban + Motorway July 2002

In many of the performance contract, both texture and roughness have been significantly improved upon,
compared with what the traditional style of contracting had delivered (Stewart, 2002). Further to this the actual
achievement under the Performance environment has further enhanced condition.

The contractor is given regular, constructive feedback on performanc e by both the network management
consultant and Transit NZ. One of the key elements of a performance contract is that the road authority and the
contractor communicate directly, openly and regularly. The performance scorecard is an annual assessment. It
is a weighted indice that takes into account the contractor’s achievements with the KPM’s, KPI’s, the network
management consultant’s score and the Visual Assessment Scorecard. The inclusion of the consultant’s
performance is an interesting concept but quite valid in that the contractor has a direct, albeit limited, influence
on the consultant’s ability to perform some of their tasks.

Discussion

Performance maintenance contracts give the contractor a number of positive opportunities to grow and improve
its business. These include but are not limited to:

• The opportunity to exercise technical and engineering abilities.

• Increase efficiencies in programming and executing physical work.
• Development of improved relationships with road authorities.
• Development of new skills and innovative ideas
• Enhanced understanding of the road authority’s needs

The road authority also accrues the following benefits from performance maintenance contracts:

• Efficiencies and innovation.

• Improvement to service delivery.
• Price certainty due to the lump sum nature of the Contract.
• Better value for money.

The performance maintenance contracts can still be improved. Consideration needs to be given to incentivise
the Contract where KPM targets are exceeded. The KPI’s need to be refined a nd honed to focus further on
service delivery enhancement in the important areas. And a more consistent approach to such contracts on a
national basis is needed.
INNOVATIONS RESULTING FROM PERFORMANCE CONTRACTS

In 1995, the first performance specificatio n implemented in New Zealand was for chip seals on state highways,
in which the contractor was responsible for designing the seal, which then had a 12 month defects liability
period. The principal performance criterion is texture depth after 12 months. In response Fulton Hogan
TM
developed Multispray , which is a computer controlled bitumen sprayer capable of varying the binder
application rate across the width of the road. Thus, less binder is sprayed in the wheelpath, where texture can be
reduced significan tly under trafficking, and more is placed between and outside the wheelpath, where the thicker
binder film is needed to hold the chips in place. At the same time, the contractor also developed a surfacing
database to collect the data needed for this perfor mance specification and produce reports the data; the data
has also been analysed to improve seal design algorithms for the variable application rate sprayer.

Another innovation driven by the requirements of performance contracts is the ultra high pressur e watercutter as
a surface treatment system for removing excess bitumen from surfaces (thereby restoring macrotexture), which
also simultaneously increases the microtexture of surface particles. The system is unique, in that it uses ultra
high pressure (36 000 psi) watercutting technology.

Transit NZ uses SCRIM (Sideways -force Coefficient Routine Investigation Machine) to measure the skid
resistance of the entire state highway network, once every year. The SCRIM is in New Zealand for only five
months every year, and additional friction testing capability is needed during that time, plus the remainder of the
year. It is uneconomical for contractors to own and operate a SCRIM, so Fulton Hogan acquired a Griptester for
friction testing. The GripTester is used for routine testing and research projects, both in response to the needs of
performance contracts and specifications.

Fulton Hogan has developed and implemented a chip seal database and management system, which is used on
chip seal designs, materials and performance. Following trials and refinement in one region, the system was
implemented nation -wide as a business tool.

The Fulton Hogan unsealed roads maintenance strategy identified the need for a. With the future trend for local
road controlling authori ties (RCA’s) towards performance contracts, the contractor needs a low cost effective tool
for roughness monitoring that quantitatively reflects the unsealed network condition and is not based on
subjective perception, which is the current situation. The O pti-Grade roughness measurement system from
Canada has been acquired, and a modified version is being used on unsealed road networks throughout New
Zealand.

SUMMARY AND CONCLUSI ONS

In summary, performance contracts are an exciting alternative to traditio nal contracting. They encourage
increased efficiencies, and allow innovation and improvements to be delivered. Modern contractors can adapt
and thrive in an environment that encourages innovation and a commercially -oriented approach to delivering a
road se rvice to the users. Relationships are key with common goals being absolutely critical to making the
performance contract a success for all.

New Zealand’s B/3 performance based specification for structural design and construction of flexible unbound
pavements was introduced to foster the use of marginal and non -conforming materials that give similar
performance to standard basecourse and sub -base materials. To date three pilot projects using B/3 have been
completed; details of two of these projects are pre sented as case studies, and the relevant outcomes of each
are discussed.

Contractors undertaking work involving performance -based specifications require highly skilled and experienced
pavement designers (either in -house or out -sourced); the road authority must also possess or have access to
experienced, knowledgeable pavement engineers in order to be able to adequately assess submitted proposals.

The benefits of performance specifications include:

• The provision of a contractual environment that encourages innovation;
• Utilising contractor’s experience;
• Focusing the client on performance rather than historical empirical relationships;
 • Apportioning the risk between client and contractor appropriately.

Road authorities and industry must collaborate, including sharing knowledge and expertise, in order for
performance contracts and specifications to be successfully introduced and implemented. These parties must
also work together to ensure that pavement research is relevant to their needs; better, more accurate and robust
(but not complex) techniques must be developed for predicting pavement performance, so that the risks for the
contractor and the road authority can be more readily quantified. In addition to potential economic advantages,
performance contracts can also preserve prime quality aggregates, so that road construction and maintenance is
achieved in an environmentally sustainable manner.

The effects of performance specifications are wide reaching and roles change significantly. The contractor is
required to possess or out -source skills such as asset management, condition assessment, pavement design
and advanced materials testing. The road authority’s role changes from designer/supervisor to that of quality
auditor.

Performance contracts, for both co nstruction and maintenance, are quite simply a better way of doing business

REFERENCES

AUSTROADS (1992) Pavement Design: Guide to Structural Design of Road Pavements . AUSTROADS,
Sydney.

AUSTROADS (2003) Summary Report: The Development of Performance Co ntracts and Specifications ,
AUSTROADS, Sydney.

Pidwerbesky, B.D. (1995) Accelerated dynamic loading of flexible pavements at CAPTIF. In Transportation
Research Record 1482 , TRB, National Research Council, Washington, D.C., pp 79 -86.

Stewart, C. (2002) H ybrid Maintenance Contracts: A Contractor’s View. New Zealand Roading Symposium ,
Auckland, New Zealand.

Transit (1995) Crushed Basecourse Aggregate. Specification M/4, Transit New Zealand, Wellington.

Transit (1997) Construction of Unbound Granular Pavem ent Courses . Specification B/2, Transit New Zealand,
Wellington, New Zealand.

Transit (2000) New Zealand Supplement to AUSTROADS Pavement Design Guide. Transit New Zealand,
Wellington.

Transit (2000a) Pilot Performance Based Specification for Structural Design and Construction of Flexible
Unbound Pavements . B/3, Transit New Zealand, Wellington.

Transit (2000b) Notes for the Evaluation of Unbound Road Base and Sub -base Aggregates . M/22, Transit New
Zealand, Wellington.

TRB (2002) Glossary of Highway Quality Assurance Terms , Transportation Research Circular E -C037, TRB,
Washington, D.C.

ACKNOWLEDGEMENTS

The author is grateful to Dave Alabaster of Transit New Zealand and Craig Stewart of Fulton Hogan for their
input to this paper.
BIOGRAPHY

Dr Brya n Pidwerbesky has been Group Technical Manager for Fulton Hogan since 2000, and is responsible for
the company’s pavement design, asset management technology, and research and development programmes.
Prior to joining Fulton Hogan, Bryan was Chief Executive of the NZ Pavement & Bitumen Contractors
Association during 1998 -2000, and Senior Lecturer in Civil Engineering at the University of Canterbury in
Christchurch NZ during 1986 -1998, where his duties included managing the CAPTIF test track.

Bryan received his doctorate in Civil Engineering fro the University of Canterbury, Christchurch, New Zealand, in
1997, and Master’s and Bachelor’s degrees in Civil Engineering from the University of Saskatchewan in1986
and 1983, respectively. He is currently on the edit orial boards of four international journals covering pavement
engineering and materials.

View publication stats