Abstract
We study resource‐constrained project scheduling problems with perturbation on activity durations. With the consideration of robustness and stability of a schedule, we model the problem as a multiobjective optimization problem. Three objectives—makespan minimization, robustness maximization, and stability maximization—are simultaneously considered. We propose a hybrid multiobjective evolutionary algorithm (H‐MOEA) to solve this problem. In the process of the H‐MOEA, the heuristic information is extracted periodically from the obtained nondominated solutions, and a local search procedure based on the accumulated information is incorporated. The results obtained from the computational study show that the proposed approach is feasible and effective for the resource‐constrained project scheduling problems with stochastic durations.
Highlights
As for its practical importance in a wide range of real-world application areas, project scheduling problems PSPs have received considerable attention in the operation research field
We propose a hybrid multiobjective evolutionary algorithm H-MOEA to solve this problem
We study a resource-constrained project scheduling problem in the presence of perturbation on activity durations
Summary
As for its practical importance in a wide range of real-world application areas, project scheduling problems PSPs have received considerable attention in the operation research field. PSPs usually address two matters: resource and time 1. The problem is modelled as a resource-constrained project scheduling problem RCPSP, which is a general type of PSPs that contains job-shop, flow-shop, and open-shop problems as special cases 2. As RCPSP is an NP-hard problem 1, heuristics or metaheuristics are preferred to solve it. For details about the models and methods of the RCPSP, readers are referred to several comprehensive surveys 3–7. It is very common that the execution of a schedule will be in the presence of uncertain factors. With the consideration of uncertainty, the problem can be modelled as stochastic
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have