Professional Documents
Culture Documents
https://doi.org/10.1007/s11227-019-03055-6
Abstract
Planning and managing resources is one of the most important topics in project
management science. Resource leveling is used for improving work efficiency and
minimizing cost throughout the life of the project. Fuzzy resource leveling mod-
els assume only truth-membership functions to deal uncertainties conditions sur-
rounded by the projects and their activities duration. In this paper, we consider the
objective function of scheduling problem is to minimize the costs of daily resource
fluctuations using the precedence relationships during the project completion time.
We design a resource leveling model based on neutrosophic set to overcome the
ambiguity caused by the real-world problems. In this model, trapezoidal neutro-
sophic numbers are used to estimate the activities durations. The crisp model for
activities time is obtained by applying score and accuracy functions. A numerical
example is developed to illustrate the validation of the proposed model in this study.
* Mohamed Abdel‑Basset
analyst_mohamed@zu.edu.eg
Mumtaz Ali
mumtaz.ali@deakin.edu.au
Asma Atef
asmaatef@fci.zu.edu.eg
1
Faculty of Computers and Informatics, Zagazig University, Sharqiyah, 44519, Egypt
2
Deakin‑SWU Joint Research Centre on Big Data, School of Information Technology, Deakin
University, Burwood, VIC 3125, Australia
13
Vol.:(0123456789)
M. Abdel‑Basset et al.
Handeled by
Resource Aggregation.
Handeled by
Resource Levelling. Time-Cost Trade offs.
Resource Allocation.
Objetive: Minimize the
Objective: Minimize the max Total Consumption over
peak or the project durations Project durations
13
Resource levelling problem in construction projects under…
late start schedules. That means for each activity in the project we can identify
the time that particular activity is going to be taking place. These activities might
require some types of resources such as man power, special type of machines or
equipment. If we want to find out the manner in which we are going to be using
that equipment, we will have to refer to these basic schedules.
From the basic schedule of activities, we can focus on this the activity resource
data. Depending upon which activities are active at what particular point of time, we
can determine the resource usage profile as shown in Fig. 2. This particular graph is
called a resource usage profile. It shows the variation in the resources which is going
to be needed to accomplish the schedule of the project.
This information is very important to the planner, the project manager, because
it would help him to identify exactly what is going to be, the resource require-
ment for the particular project and exactly how much resource is required during
what particular periods of time. The project manager can make his provisions
for hiring and firing appropriately. Renewable resources are basically handled
through three different mechanisms. The simplest way of handling these resources
is through what we call resource aggregation. Aggregation, simply means as the
name suggests, is the process of working out the resource usage profile from the
Resource
Usage
Schedule
8
Resource Usage Profile
T1 T2 T3 TN
Time
Fig. 2 Resource profile
13
M. Abdel‑Basset et al.
schedule. Resource aggregation is actually simply trying to find out the resource
requirements for a particular schedule. It does not in any way try to influence
those resource requirements. It is simply like taking a snapshot of a particular
schedule as shown in Fig. 2. On the other hand, resource leveling is more detailed
in the sense that it looks at the resource profile and we might not find the resource
profile very satisfactory. It might want to change the schedule to get a better more
acceptable resource profile as shown in Fig. 3, by minimizing the max peak and
also minimize the fluctuates as much as possible while make the completion time
of the project constant. Essentially, that is the process of resource leveling and
limited resource allocation is the process of trying to conform to resource avail-
abilities and trying to minimize the project duration as shown in Fig. 4. There are
procedures available for resource leveling. There are both analytical procedures
as well as heuristic procedures but the difficulty with the analytical procedures is
that they are capable of handling only a small number of jobs or very small pro-
jects and for real-life projects it is not possible to do resource leveling for a large
project which involves hundreds of activities. For this reason, generally, resort
is made to heuristic procedures. Although heuristic procedures do not guarantee
optimal solutions nevertheless, they are practical procedures which can be used
for this particular situation.
There are many studies [1–11] discussed the traditional procedures for leveling
resources. Many heuristics and meta-heuristics studies [12–23] also discussed the
resource leveling problem. Many fuzzy models [24–27] are also study the prob-
lem of resource leveling under uncertain conditions. In this paper, we will use
neutrosophic heuristic procedure for resource leveling.
R (T) Original
Resource Profile
Obtained
Resource Profile
T Time
Fig. 3 Resource leveling
13
Resource levelling problem in construction projects under…
Original
R (T) Resource Profile
Best Leveled
Resource Profile
Resource
Allocation Profile
T TN TI Time
Fig. 4 Resource allocation
2 Preliminaries
Definition 1 [31] Let be a space of points and x ∈ X . A neutrosophic set A in X
is defined by a truth-membership function TA(x), an indeterminacy-membership
function IA(x) and a falsity-membership function FA(x), TA(x), IA(x) and FA(x) are
real standard or real nonstandard subsets of [− 0, 1+]. That is TA(x):→ [− 0, 1 +] IA
(x):→[− 0 1 +] and FA(x):→[− 0, 1 +]. There is no restriction on the sum of TA(x), IA
(x) and FA(x), so 0 −≤ sup (x) + sup ≤ 3 +.
Definition 2 [30, 31, 32] Let X be a universe of discourse. A single valued neutro-
sophic set A over X is an object taking the form A = {〈x, TA(x), IA(x), FA(x),〉:x ∈ X},
where TA(x):X→ [0, 1], IA(x):X→ [0, 1] and FA(x):X→[0, 1] with 0 ≤ TA(x) + IA
(x) + FA(x) ≤ 3 for all x X. The intervals TA(x), IA(x) and FA(x) represent the truth-
membership degree, the indeterminacy-membership degree and the falsity-mem-
bership degree of x to A, respectively. For convenience, a single valued neu-
trosophic (SVN) number is represented by A = (a, b, c), where a, b, c [0, 1] and
a+b+c ≤ 3.
13
M. Abdel‑Basset et al.
Definition 3 [34] Suppose 𝛼ã , 𝜃ã , 𝛽ã ∈ [0,1] and a1 , a2 , a3 , a4 𝜖 R, where
a1 ≤ a2 ≤ a3 ≤ a4 . Then, a single valued trapezoidal neutrosophic number ã = 〈(a1 ,
a2 , a3 , a4 ); 𝛼ã , 𝜃ã , 𝛽ã 〉 is a special neutrosophic set on the real line set R, whose
truth-membership, indeterminacy membership and falsity-membership functions are
defined as:
� �
⎧ 𝛼 x−a1 (a1 ≤ x ≤ a2 )
⎪ ã a2 −a1
⎪𝛼 (a2 ≤ x ≤ a3 )
Tã (x) = ⎨ ã � a −x � (1)
⎪ 𝛼ã a4 −a3
4
(a3 ≤ x ≤ a4 )
⎪0 otherwise
⎩
where 𝛼ã , 𝜃ã and 𝛽ã typify the maximum truth-membership degree, the minimum
indeterminacy-membership degree and the minimum falsity-membership degree,
respectively. A single valued trapezoidal neutrosophic number ã = 〈(a1 , a2 , a3 , a4 );
𝛼ã , 𝜃ã , 𝛽ã 〉 may express an ill-defined quantity of the range, which is approximately
equal to the interval [ a2 , a3].
Resource aggregation, resource leveling and resource allocations are three ways for
handling the renewable resources. Resource aggregation is just construct resource
usage profile from the project schedule and the activities resource data. Resource
allocation aimed to minimize the project duration with fixed resource availabilities.
Resource leveling, the core of this research, aimed to smooth resource usage profile
by shifting the tasks relative to their available slack or floats without worsening the
13
Resource levelling problem in construction projects under…
duration of the project. There are different criteria affected leveling resources, such
as the peak level of the resources usage, as used in workload smoothing heuristic
developed by Levy, Thampson and Wiest [28]; another criteria that may affect the
resource leveling is the sum of squares of resources usage as used by Burgess and
Killebrew heuristic [29]. For more reality, we present the proposed Neutrosophic—
Burgess heuristic algorithm for the purpose of best resource leveling to minimize
the fluctuations in day to day of resources used during the project execution time in
uncertain situations as follows:
13
M. Abdel‑Basset et al.
Table 1 Project data
Activity Activity time (days) as trapezoidal neutrosophic Predecessors
4 Illustrative example
13
Resource levelling problem in construction projects under…
as given in Table 2, and then we make the initial early start scheduling resource
profile as shown in Fig. 6, that obtained from the bar chart in Fig. 5 with the list
of activities resource, and then calculate the total resource requirements sum of
squares for each time unit as given in Table 3.
Then starting with activity (I), and reschedule it at different times according to
the available total floats on it. Science activity (I) has 5 units of floats, activity (I)
will be rescheduled five times from activity (I) with one unit of float (I1) to (I5)
with five units of float (the max number of available floats on activity I), then cal-
culate the total summation of squares of resource requirements at different float
times and select the schedule that give the lowest total sum of squares of resource
requirements for each time unit. Calculations of resource requirements to the dif-
ferent floats on activity (I) are given in Table 4.
According to the Neutrosophic—Burgess algorithm, holding activity (I) fixed
with the early start schedule of (628) as a minimum sum of squares of resource
requirements. Taking activity (H) in the next steps of our algorithm and also
reschedule it due to the available floats on activity (H) from H1 to H4 to give the
lowest total summation of squares of resources requirements as given in Table 5.
A (1)
A
B (4)
B
C (2)
C
D (4)
D
E (1)
E
F (3)
F
G (2)
G
H (3)
H
I (3)
I
J (2)
J
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
13
13
Table 3 Resource requirements for each time unit and total sum of square of resource requirements
Time per Resource required per day
t=1 R2
∑15
day (T)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day (10) Day (11) Day (12) Day (13) Day (14) Day (15)
Resource 7 9 12 12 9 6 4 4 4 5 2 2 2 2 2 628
Required
(R)
M. Abdel‑Basset et al.
Table 4 Resource requirements of rescheduling activity (I) at different float times
Activity Resource requirements per day
t=1 R2
∑15
(I)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
I1 7 9 12 12 9 6 1 4 4 5 5 2 2 2 2 634
Resource levelling problem in construction projects under…
I2 7 9 12 12 9 6 1 1 4 5 5 5 2 2 2 640
I3 7 9 12 12 9 6 1 1 1 5 5 5 5 2 2 646
I4 7 9 12 12 9 6 1 1 1 2 5 5 5 5 2 646
I5 7 9 12 12 9 6 1 1 1 2 2 5 5 5 5 646
13
13
Table 5 Resource requirements of rescheduling activity (H) at different float times
Activity Resource requirements per day
t=1 R2
∑15
(H)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
H1 7 9 9 12 9 9 4 4 4 5 2 2 2 2 2 610
H2 7 9 9 9 9 9 7 4 4 5 2 2 2 2 2 580
H3 7 9 9 9 6 9 7 7 4 5 2 2 2 2 2 568
H4 7 9 9 9 6 6 7 7 7 5 2 2 2 2 2 556
M. Abdel‑Basset et al.
Table 6 Resource requirements of rescheduling activity (H) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(I)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
I1 7 9 9 9 6 6 4 7 7 5 5 2 2 2 2 544
Resource levelling problem in construction projects under…
I2 7 9 9 9 6 6 4 4 7 5 5 5 2 2 2 532
I3 7 9 9 9 6 6 4 4 4 5 5 5 5 2 2 520
I4 7 9 9 9 6 6 4 4 4 2 5 5 5 5 2 520
I5 7 9 9 9 6 6 4 4 4 2 2 5 5 5 5 520
13
13
Table 7 Resource requirements of rescheduling activity (G) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(G)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
G1 7 9 7 9 6 6 6 4 4 2 2 5 5 5 5 508
G2 7 9 7 7 6 6 6 6 4 2 2 5 5 5 5 496
G3 7 9 7 7 4 6 6 6 6 2 2 5 5 5 5 496
G4 7 9 7 7 4 4 6 6 6 4 2 5 5 5 5 488
G5 7 9 7 7 4 4 4 6 6 4 4 5 5 5 5 480
M. Abdel‑Basset et al.
Resource levelling problem in construction projects under…
13
13
Table 8 Resource requirements of rescheduling activity (I) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(I)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
I1 7 9 7 7 4 4 4 9 9 7 7 2 2 2 2 552
I2 7 9 7 7 4 4 4 6 9 7 7 5 2 2 2 528
I3 7 9 7 7 4 4 4 6 6 7 7 5 5 2 2 504
I4 7 9 7 7 4 4 4 6 6 4 7 5 5 5 2 492
I5 7 9 7 7 4 4 4 6 6 4 4 5 5 5 5 480
M. Abdel‑Basset et al.
Table 9 Resource requirements of rescheduling activity (F) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(F)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
F1 7 9 4 7 4 4 7 6 6 4 4 5 5 5 5 480
Resource levelling problem in construction projects under…
F2 7 9 4 4 4 4 7 9 6 4 4 5 5 5 5 492
F3 7 9 4 4 1 4 7 9 9 4 4 5 5 5 5 522
F4 7 9 4 4 1 1 7 9 9 7 4 5 5 5 5 540
F5 7 9 4 4 1 1 4 9 9 7 7 5 5 5 5 540
13
13
Table 10 Resource requirements of rescheduling activity (B) at different float times
Activity Resource required per days
t=1 R2
∑15
(B)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
B1 3 9 8 7 4 4 7 6 6 4 4 5 5 5 5 488
B2 3 5 8 11 4 4 7 6 6 4 4 5 5 5 5 543
B3 3 5 4 11 8 4 7 6 6 4 4 5 5 5 5 504
B4 3 5 4 7 8 8 7 6 6 4 4 5 5 5 5 480
M. Abdel‑Basset et al.
Table 11 Resource requirements of rescheduling activity (A) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(A)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
A1 2 5 5 7 8 8 7 6 6 4 4 5 5 5 5 484
Resource levelling problem in construction projects under…
A2 2 4 5 8 8 8 7 6 6 4 4 5 5 5 5 490
A3 2 4 4 8 9 8 7 6 6 4 4 5 5 5 5 498
A4 2 4 4 7 9 9 7 6 6 4 4 5 5 5 5 500
A5 2 4 4 7 8 9 8 6 6 4 4 5 5 5 5 498
13
13
Table 12 Resource requirements of rescheduling activity (F) at different float times
Activity Resource requirements per days
t=1 R2
∑15
(F)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day Day Day Day Day Day
(10) (11) (12) (13) (14) (15)
F1 3 5 4 7 8 8 7 6 6 4 4 5 5 5 5 480
F2 3 5 4 4 8 8 7 9 6 4 4 5 5 5 5 492
F3 3 5 4 4 5 8 7 9 9 4 4 5 5 5 5 498
F4 3 5 4 4 5 5 7 9 9 7 4 5 5 5 5 492
F5 3 5 4 4 5 5 4 9 9 7 7 5 5 5 5 492
M. Abdel‑Basset et al.
Table 13 Resource requirements for each time unit and total sum of square of resource requirements after applying the proposed procedure
Time per Resource requirements per days
t=1 R2
∑15
day (T)
Day (1) Day (2) Day (3) Day (4) Day (5) Day (6) Day (7) Day (8) Day (9) Day (10) Day (11) Day (12) Day (13) Day (14) Day (15)
Resource levelling problem in construction projects under…
Resource 3 5 4 7 8 8 7 6 6 4 4 5 5 5 5 480
required
(R)
13
M. Abdel‑Basset et al.
R (T)
12
11
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
T
proposed procedure, the max peak is reduced from 12 to 8, and the fluctuation in
day to day of resources used is also minimized as the purpose of best resource lev-
eling. Also the sum of squares of resources requirements is reduced from 628 to 480
as given in Tables 3, and 13, before and after applying the new proposed method.
This indicates the effectiveness of the proposed procedure for leveling resources in
the absence of accuracy and confirmation of the times of project activities.
Neutrosophic set is the most comprehensive set, which includes both fuzzy set and
intuitionistic fuzzy set, as it considers the indeterminacy function in addition to
truth-membership and falsity membership, being suitable in analyzing real situa-
tions. Also, in real-life situations, accurate judgments are rarely since ambiguity and
uncertainty surround the decision-making process. Resource leveling is one of the
mechanisms used for handling the renewable resources. In this research, we proposed
Neutrosophic—Burgess heuristic algorithm for the purpose of best resource leveling
to minimize the fluctuations in day-to-day resource usage during the project. When
we compare the resource leveling under the neutrosophic interval method with the
resource leveling under the traditional and fuzzy methods, we find that the most ade-
quate and effective method that provide the parameters under uncertainty and also
13
Resource levelling problem in construction projects under…
R (T)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 T
provide range of parameter in the indeterminacy interval with more real conditions.
This work is done based on trapezoidal neutrosophic number. The numerical exam-
ple indicates the effectiveness of the proposed procedure for leveling resources in the
absence of accuracy and confirmation of the times of project activities. But in the
future, we plan to use the heptagonal neutrosophic number and pentagonal neutro-
sophic number. We also plan to use the neutrosophic model for handling the non-
renewable resource to minimize the cost trade-offs, and also we use the neutrosophic
model for best resource allocation problem as an approach for handling the renewable
resource in the case of limited resources are available. Also we can use neutrosophic
theory for scheduling resource-constrained project problem, and scheduling multi-
mode resource-constrained project problems based on neutrosophic environment.
Conflict of interest Authors declare that there is no conflict of interest about the research.
References
1. Leu SS, Hung TH (2002) An optimal construction resource leveling scheduling simulation model.
Can J Civ Eng 29(2):267–275
13
M. Abdel‑Basset et al.
13
Resource levelling problem in construction projects under…
30. Saaty TL (2006) The analytic network process. In: Decision making with the analytic network pro-
cess, Springer, Boston, pp 1–26
31. Abdel-Baset M, Hezam IM, Smarandache F (2016) Neutrosophic goal programming. Neutrosophic
Sets Syst 11:112–118
32. Mahdi IM, Riley MJ, Fereig SM, Alex AP (2002) A multi-criteria approach to contractor selection.
Eng Constr Architect Manag 9(1):29–37
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
13