Resource levelling in projects considering different activity execution modes and splitting

Abstract

Purpose This study aims to develop a mathematical programming model for preemptive multi-mode resource-constrained project scheduling problems in construction with the objective of levelling resources considering renewable and non-renewable resources. Design/methodology/approach The proposed model was solved by the exact method and the genetic algorithm integrated with the solution modification procedure coded with MATLAB software. The Taguchi method was applied for setting the parameters of the genetic algorithm. Different numerical examples were used to show the validation of the proposed model and the capability of the genetic algorithm in solving large-sized problems. In addition, the sensitivity analysis of two parameters, including resource factor and order strength, was conducted to investigate their impact on computational time. Findings The results showed that preemptive activities obtained better results than non-preemptive activities. In addition, the validity of the genetic algorithm was evaluated by comparing its solutions to the ones of the exact methods. Although the exact method could not find the optimal solution for large-scale problems, the genetic algorithm obtained close to optimal solutions within a short computational time. Moreover, the findings demonstrated that the genetic algorithm was capable of achieving optimal solutions for small-sized problems. The proposed model assists construction project practitioners with developing a realistic project schedule to better estimate the project completion time and minimize fluctuations in resource usage during the entire project horizon. Originality/value There has been no study considering the interruption of multi-mode activities with fluctuations in resource usage over an entire project horizon. In this regard, fluctuations in resource consumption are an important issue that needs the attention of project planners.

Time-Cost Tradeoff and Resource-Scheduling Problems in Construction: A State-of-the-Art Review

“…Khalilzadeh [28] presented a model to level resources by using mathematical programming considering renewable and non-renewable resources for preemptive multi-mode RCPS in construction through using the exact method and the GA integrated with the solution modification. As the exact method failed to obtain the optimal solution for large-size problems, GA obtains the nearest optimal solutions within reasonable and short computational duration.…”

“…The model aids construction project practitioners in creating a realistic and practical project schedule to better calculate the project duration and reduce fluctuations in resource utilization during the project. The major gaps of the model developed by Khalilzadeh [28] that assuming all activities can be interrupted and the activity splitting cost is neglected. Also, excluding project cost and quality could be important gaps.…”

“…Unconstrained resources problem is attempting to balance resource use across the whole project life cycle [6], in different word, achieving better resource distribution (histogram) through the project execution especially, labors and equipment. The major issues result from unleveled resources; shortage, idle time and hiring and firing, which lead to extra costs, delaying project and sometimes negative social effects [28]. So, some researchers keen on solving this problem [6], [25]- [28].…”

“…The major issues result from unleveled resources; shortage, idle time and hiring and firing, which lead to extra costs, delaying project and sometimes negative social effects [28]. So, some researchers keen on solving this problem [6], [25]- [28].…”

Robust forensic-based investigation algorithm for resource leveling in multiple projects

“…The purpose of resource levelling is to reduce resource usage fluctuations as much as possible in time span along the project timelines. Resource levelling attempts to schedule noncritical activities within their available floats without changing total project time to attain a good resource histogram [10]. The problems of resource leveling can be classified into four categories: (1) considering a single resource in a project [11], (2) handling multiple types of resources in a project [12], (3) dealing with a single resource in many projects [13], and (4) leveling multiple resources in different projects [14][15][16].…”