Activity scheduling and resource allocation with uncertainties and learning in activities
| Pages | 1289-1320 |
| Date | 08 July 2019 |
| DOI | https://doi.org/10.1108/IMDS-01-2019-0002 |
| Published date | 08 July 2019 |
| Author | Felix T.S. Chan,Zhengxu Wang,Yashveer Singh,X.P. Wang,J.H. Ruan,M.K. Tiwari |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
Activity scheduling and resource
allocation with uncertainties and
learning in activities
Felix T.S. Chan
Department of Industrial and Systems Engineering,
Hong Kong Polytechnic University,
Kowloon, Hong Kong
Zhengxu Wang
Department of Business Administration,
Dongbei University of Finance and Economics, Dalian, China
Yashveer Singh
Department of Industrial and Systems Engineering,
Indian Institute of Technology Kharagpur, Kharagpur, India
X.P. Wang
Institute of Systems Engineering,
Dalian University of Technology, Dalian, China
J.H. Ruan
College of Economics and Management,
Northwest Agriculture and Forestry University,
Yangling, China, and
M.K. Tiwari
Department of Industrial and Systems Engineering,
Indian Institute of Technology Kharagpur,
Kharagpur, India
Abstract
Purpose –The purpose of this paper is to develop a model which schedules activities and allocates resources
in a resource constrained project management problem. This paper also considers learning rate and
uncertainties in the activity durations.
Design/methodology/approach –An activity schedule with requirements of different resource units is
used to calculate the objectives: makespan and resource efficiency. A comparisons between non-dominated
sorting genetic algorithm –II (NSGA-II) and non-dominated sorting genetic algorithm –III (NSGA-III) is done
to calculate near optimal solutions. Buffers are introduced in the activity schedule to take uncertainty into
account and learning rate is used to incorporate the learning effect.
Findings –The results show that NSGA-III gives better near optimal solutions than NSGA-II for multi-
objective problem with different complexities of activity schedule.
Research limitations/implications –The paper does not considers activity sequencing with multiple
activity relations (for instance partial overlapping among different activities) and dynamic events occurring
in between or during activities.
Practical implications –The paper helps project managers in manufacturing industry to schedule the
activities and allocate resources for a near-real world environment.
Industrial Management & Data
Systems
Vol. 119 No. 6, 2019
pp. 1289-1320
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-01-2019-0002
Received 1 January 2019
Revised 8 March 2019
Accepted 16 April 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
The work describedin this paper was supportedby grant from The Natural Science Foundationof China
(Grant No. 71471158), China Postdoctoral Science Foundation funded project (Grant No. 2018M631792)
and The Department of Education of Liaoning Province (Grant No. LN2017QN006).
1289
Activity
scheduling and
resource
allocation
Originality/value –This paper takes into account both the learning rate and the uncertainties in the activity
duration for a resource constrained project management problem. The uncertainty in both the individual
durations of activities and the whole project duration time is taken into consideration. Genetic algorithms
were used to solve the problem at hand.
Keywords Critical chain buffers, Evolutionary multi-objective optimization, Learning effect
Paper type Research paper
1. Introduction
Scheduling is basically the allocation of finite resources to the various activities to attain
optimality. The Resource Constrained Project Scheduling Problem (RCPSP) is considered to be a
NP-hard problem, but it can be readily stated as a problem with renewable resources and
activities, with the aim of minimising the duration of the activities (Brucker, 2002). The
renewable resources are machines, equipment, manpower, etc. There are a lot of risks involved in
any industry, and the major factors that account for risk or uncertainty are the non-availability of
raw materials, longer activity time than usual and the reprocessing of defective items.
1.1 Managerial challenges
Apart from the technical difficulties in any complex project, other issues arise due to managerial
complexity. So, it is necessary to coordinate among different activities. In any industry,
management competency, i.e. scheduling and planning plays a significant role. A process can
be completed but the main aim of the manager is to schedule activities in such a way so as to
decrease the makespan and at the same time allocating resources efficiently. Scheduling the
sequence of activities is the main challenge that project managers face in project planning. The
critical chain (Rand, 2000) and the program evaluation and review technique (Vanhoucke, 2016;
Malcolm et al., 1959) are techniques that the managers use for scheduling a project.
The managerial challenges that the project managers have to face (Xiong et al., 2016) in
scheduling a project are: scheduling the activities and allocating the resources. The project
manager must decide on the number of resources that should be given to each activity and
simultaneouslydecide on the sequence of each activity.Deciding the sequence of each activity
is the key processto find the makespan of the whole productionprocess; duration of uncertain
activities. There are many factors that affect the duration of activities, and predicting the
finishing timeof each activity is very difficult, thus the actualactivity duration can be known
only after the activity is completed. The uncertainty can arise due to various unavoidable
reasons likedamage to machinery or the machinetaking more time than usualto complete the
desired task;time dependency. With the increasein experience in undertakingan activity, the
time requiredto complete the activity also decreasesand, moreover, the technologicalrisk also
decreases with hands-on experience. When the operators gain experience in working on a
particular machine, they then tend to take less time than usual to complete the required
activity; conflicting objectives. Just like any other multi-objective complex project, the
objective functions tend to conflict with each other. So, we must keep a balance in solving a
complex problemthat has multiple objectivefunctions, while trying to keep thevalue of all the
objective functions in the desired range.
1.2 Uncertainty in projects
Apart from the uncertainty in durations of each activity, there can also be uncertainty in the
completion of the whole project because of some undesirable factors like workers strikes, natural
calamities, machine breakdowns, etc. In order to implement the schedule of the project as
planned and to save the critical chain, buffers are introduced in the project planning (Goldratt,
1997). As mentioned by Patrick (1999), buffers in scheduling are decided in a different manner
from the management of slack. Due to its random occurrence in any project schedule slack is
ineffective in reducing the effects of variations on the total project duration (Yu et al., 2018);
whereas, buffers consider the uncertainty in the estimates of the activity durations.
1290
IMDS
119,6
1.3 Advancement in knowledge
The amount of work that has been put in to help project managers with these mentioned
challenges is minimal (Xiong et al., 2014). So, to help the managers to schedule the activities
and allocate the resources in a better way, a good model is necessary. This, in turn, will
decrease the makespan of production, increase the efficiency of resources and increase the
profit for the company. This paper describes a new model for RCPSP with multiple
objectives, with the number of renewable resources available as constraints. The
uncertainty in both the individual durations of activities and the whole project duration time
is taken into consideration. This model incorporates the learning rate in activities, which
results in a decrease in the activity durations with the gain in the experience of workers.
Critical chain buffers are incorporated in the activity sequence to reduce the effect of
uncertainties in production. The main objectives of this paper are: minimisation of
makespan and maximisation of resource efficiencies; a comparison of the results between
two very popular algorithms: NSGA-II and NSGA-III. Like any other project planning
problem, our approach also follows a right-skewed βdistribution to calculate the duration of
each activity.
1.4 Sectional contents
Section 1 consists ofthe introduction to the topic, Section 2 describes the literature review of
the work done, Section 3 describes the problem statement that is to be solved, Section 4
describes the algorithm used for solving the present problem, Section 5 presents the results
obtained fromthe problem statement. Section6 gives the conclusions and future relatedwork.
2. Literature review
Many researchers have worked in the field of project management but the area of resource
efficiency is still new. Resources (Kolisch and Padman, 2001) are classified into: renewable,
non-renewable, partially renewable and doubly constrained. RCPSP consists of renewable
resource constraints with precedence constraints for the activities. RCPSP with heuristic
and meta-heuristic approaches has been reported in the literature (Herroelen et al., 1998;
Kolisch and Padman, 2001).
2.1 Uncertainty in activities
The main problem with project scheduling arisesbecause of uncertainty in the completionof
an activity. Uncertainties in production may cause serious disruption in schedules (Koh and
Gunasekaran, 2006; Chakrabortty et al., 2017). These uncertainties may arise from various
different sources: activity disruptions (activities may take less or more time than usual
estimated duration), resource disruptions (the resources may not be available for use due to
some faults),due dates and ready times may have to be changed, latearrival of raw materials,
delays due to weather conditions, etc. (Herroelen and Leus, 2005). The existing research in
project scheduling considersactivity disruptions (Lamas and Demeulemeester, 2016;Ma et al.,
2016; Artigueset al., 201 3; Bruni et al., 2011;Deblaere et al., 2011a; Van de Vonder et al., 2008),
resource disruptions (Chakrabortty et al., 2016; Lambrechts et al., 2008, 2011) or eve n both
resource and activity disruptions (Fu et al., 2015; Deblaere et al., 2011b). Although there are
many variations in project scheduling from different sources, the main objectives of activity
scheduling pertain to the beginning and/orend times of the activities and the most important
scheduling objective is makespan (Artigues et al.,2013).
2.2 Learning effect on operators
The phenomenon of learning rate has received a lot of consideration by project managers in
scheduling (Azzouz et al., 2018). The effect of learning rate in scheduling was first
1291
Activity
scheduling and
resource
allocation
To continue reading
Request your trial