This paper presents an improvement cuckoo search (ICS) algorithm for minimizing the project duration of the resource-constrained project scheduling problem with generalized precedence relations (RCPSP/max). The ICS algorithm is designed as follows, each individual is coded based on the priority of activity for scheduling as to be in a form which is suitable for matching the characteristics of Lévy flight, and the individual should be resolved by serial schedule generation scheme with the improved double justification. To avoid low efficiency and local minimum, an elite individual local search method is adopted in CS. Finally, computational results in the RCPSP/max benchmarks show the efficiency of the proposed algorithm.
Download Full PDF Version (Non-Commercial Use)