Modeling of Ant Colony's Labor Division for the Multi-Project Scheduling Problem and Its Solution by PSO

For Multi-mode resource constrained multi-project scheduling problem (MRCMPSP), the solutio largely depends on the task modes and task scheduling schemes. Besides, the randomness of task scheduling makes it more difficult to solve the problem. Due to the precedence and resource constraints, the optimal solutions or feasible solutions might be found in different feasible regions, requiring an algorithm with strong global and local search ability. Considered the features of the MRCMPSP problem, ant colony's labor division is employed to establish a task priority scheduling model firstly. Then, improved particle swarm optimization (PSO) is used to find out the optimumn scheduling scheme. The approach integrating the above two algorithms has abilities of both local search and global search. Finally, the numerical experiment is employed and the results verify the feasibility and effectiveness of the proposed approach.