Heuristic procedures for probabilistic project scheduling (original) (raw)
SSRN Electronic Journal, 2014
The purpose of this research is to develop a new procedure for generating a proactive baseline schedule for the resource constrained project scheduling problem. The main advantage of this new procedure is that it is completely independent of the reactive policy applied. This contrasts with the traditional methods that assume a predefined reactive policy. First, we define a new robustness measure, then we introduce a branchand-cut method for solving a sample average approximation of our original problem. In a computational experiment, we show that our procedure outperforms two other published methods, assuming different robustness measures.
Robust optimization for resource-constrained project scheduling with uncertain activity durations
Flexible Services and Manufacturing Journal, 2013
The purpose of this paper is to propose models for project scheduling when there is considerable uncertainty in the activity durations, to the extent that the decision maker cannot with confidence associate probabilities with the possible outcomes of a decision. Our modeling techniques stem from robust optimization, which is a theoretical framework that enables the decision maker to produce solutions that will have a reasonably good objective value under any likely input data scenario. We develop and implement a scenario-relaxation algorithm and a scenario-relaxation-based heuristic. The first algorithm produces optimal solutions but requires excessive running times even for medium-sized instances; the second algorithm produces high-quality solutions for medium-sized instances and outperforms two benchmark heuristics.
Predetermined intervals for start times of activities in the stochastic project scheduling problem
Annals of Operations Research, 2010
This paper proposes a new methodology to schedule activities in projects with stochastic activity durations. The main idea is to determine for each activity an interval in which the activity is allowed to start its processing. Deviations from these intervals result in penalty costs. We employ the Cross-Entropy methodology to set the intervals so as to minimize the sum of the expected penalty costs. The paper describes the implementation of the method, compares its results to other heuristic methods and provides some insights towards actual applications.
Resource Constrained Project Scheduling with Stochastic Resources
2018
Despite the dynamic nature of real life scheduling problems, few studies focus on stochastic resource constrained project scheduling problem and its variants. In this study, we consider stochastic resource possibilities and propose a new chance constraint, piecewise-linear and mixed integer programming model. Model is tested and verified with known project instances. One of the main strengths of the proposed model is it can be used to construct baseline schedules with a predetermined confidence interval. This gives scheduler an opportunity to construct proactive actions in order to minimize disruptions.
Timeslack-Based Techniques for Generating Robust Project Schedules Subject to Resource Uncertainty
SSRN Electronic Journal, 2000
The classical, deterministic resource-constrained project scheduling problem has been the subject of a great deal of research during the previous decades. This is not surprising given the high practical relevance of this scheduling problem. Nevertheless, extensions are needed to be better able to cope with situations arising in practice such as multiple activity execution modes, activity duration changes and resource breakdowns. In this paper we analytically determine the * impact of unexpected resource breakdowns on activity durations. Furthermore, using this information we develop an approach for inserting explicit idle time into the project schedule in order to protect it as well as possible from disruptions caused by resource unavailabilities. This strategy will be compared to a traditional simulation-based procedure and to a heuristic developed for the case of stochastic activity durations.
Journal of Scheduling, 2008
Research concerning project planning under uncertainty has primarily focused on the stochastic resource-constrained project scheduling problem (stochastic RCPSP), an extension of the basic RCPSP, in which the assumption of deterministic activity durations is dropped. In this paper, we introduce a new variant of the RCPSP for which the uncertainty is modeled by means of resource availabilities that are subject to unforeseen breakdowns. Our objective is to build a robust schedule that meets the project due date and minimizes the schedule instability cost, defined as the expected weighted sum of the absolute deviations between the planned and actually realized activity starting times during project execution. We describe how stochastic resource breakdowns can be modeled, which reaction is recommended when a resource infeasibility occurs due to a breakdown and how one can protect the initial schedule from the adverse effects of potential breakdowns.
Proactive Resource Allocation Heuristics for Robust Project Scheduling
SSRN Electronic Journal, 2000
The well-known deterministic resource-constrained project scheduling problem (RCPSP) involves the determination of a predictive schedule (baseline schedule or pre-schedule) of the project activities that satisfies the finish-start precedence relations and the renewable resource constraints under the objective of minimizing the project duration. This pre-schedule serves as a baseline for the execution of the project. During execution, however, the project can be subject to several types of disruptions that may disturb the baseline schedule. Management must then rely on a reactive scheduling procedure for revising or reoptimizing the pre-schedule.
Springer eBooks, 2014
The purpose of this paper is to propose models for project scheduling when there is considerable uncertainty in the activity durations, to the extent that the decision maker cannot with confidence associate probabilities with the possible outcomes of a decision. Our modeling techniques stem from robust optimization, which is a theoretical framework that enables the decision maker to produce solutions that will have a reasonably good objective value under any likely input data scenario. We develop and implement a scenario-relaxation algorithm and a scenario-relaxation-based heuristic. The first algorithm produces optimal solutions but requires excessive running times even for medium-sized instances; the second algorithm produces high-quality solutions for medium-sized instances and outperforms two benchmark heuristics.
Time slack-based techniques for robust project scheduling subject to resource uncertainty
Annals of Operations Research, 2011
The resource-constrained project scheduling problem (RCPSP) has been the subject of a great deal of research during the previous decades. This is not surprising given the high practical relevance of this scheduling problem. Nevertheless, extensions are needed to be able to cope with situations arising in practice such as multiple activity execution modes, activity duration changes and resource breakdowns. In this paper we analytically determine the impact of unexpected resource breakdowns on activity durations. Furthermore, using this information we develop an approach for inserting explicit idle time into the project schedule in order to protect it as well as possible from disruptions caused by resource unavailabilities. This strategy will be compared to a traditional simulation-based procedure and to a heuristic developed for the case of stochastic activity durations.
The Stochastic Resource-Constrained Project Scheduling Problem
Springer eBooks, 2014
Resource Constrained Project Scheduling Problem (RCPSP) is a well-known scheduling problem where aim is to optimize an objective under limited resources and activity constraints. From the real-life perspective, it has many applications such as construction, manufacturing, and R&D projects. It is shown by Blazewicz et al. [1] that RCPSP is NP-hard in the strong sense. Due to the nature of the problem itself, nature inspired algorithms are used extensively for the solution of the problem. Intelligent systems based on such algorithms can be used effectively if the proposed models can cover real life problems' complexities. Therefore, intelligent systems should be designed based on best fit models. Mainly RCPSP is modeled and solved in a deterministic environment where parameters are all assumed to be known [2]. Real life projects are consistent; production attributes are stochastic [3], and parameters are subject to change during execution of a project. Scheduling real life problems are subject to considerable uncertainties due to the dynamic nature of project environment [4]. These uncertainties and fluctuations may stem from project itself such as activity completion times, resource estimates, material delivery dates project externalities like severe weather conditions, owner's scope changes or imposed deadline changes. Thus, the limits of deterministic models are criticized by several researchers [5]. Contrary to the deterministic models, stochastic models portray the dynamic project environment with the assumption of varying project parameters. For the stochastic RCPSP, few researchers tried to model activity disruptions [6-7] and resource fluctuations separately [5]. Basic idea is to construct a