A Hybrid Metaheuristic for Project Scheduling Problems with Fuzzy Activity Durations to support the Ös-Dráva Water Management Programme

Abstract

Spatial developmental programs in Hungary are important to support economically lagging regions. As a result of their nature they can be considered as big projects, with long timescales, high levels of capital, causing significant impact on the targeted field. For an effective performance, an accurate planning phase is essentially important. In this paper we concentrate on the project scheduling element of planning, to reduce risk and gain knowledge on the possible, net present value and makespans. The Ős-Drava program is a water-management based spatial development program. There have been no reference projects like this in Hungary that can support the scheduling of the problem with applicable information. This means that a highly uncertain case is considered with resource constraints implied. Our goal is to support the programme with an appropriate project scheduling method, resulting in a reliable makespan. For this we present a stochastic programming model and a hybrid metaheuristic for resource constrained project scheduling problems. The stochastic programming model was presented by Goel and Grossman[1], considering the net present value of subprojects. The hybrid metaheuristic was presented by Danka [2] which is a modification of the model developed by Csebfalvi [3]. In the presented approach, it is assumed that each activity duration is considered as a fuzzy variable which, can be described with an appropriate membership function. The heuristic algorithm produces “robust” resource-feasible schedules which are totally immune against uncertainties in the activity durations. The hybrid algorithm presented is based on the “Sounds of Silence” harmony search metaheuristic developed by Csebfalvi et al. [4, 5].