Published online by Cambridge University Press: 11 July 2014
Many real-world scheduling problems can be modeled as Multi-mode Resource ConstrainedProject Scheduling Problems (MRCPSP). However, the MRCPSP is a strong NP-hard problem andvery difficult to be solved. The purpose of this research is to investigate a moreefficient alternative based on ant algorithm to solve MRCPSP. To enhance the generalityalong with efficiency of the algorithm, the rule pool is designed to manage numerouspriority rules for MRCPSP. Each ant is provided with an independent thread and endowedwith the learning ability to dynamically select the excellent priority rules. In addition,all the ants in the ant algorithm have the prejudgment ability to avoid infeasible routesbased on the branch and bound method. The algorithm is tested on the well-known benchmarkinstances in PSPLIB. The computational results validate the effectiveness of the proposedalgorithm.