Modeling a bi-objective multi-skilled project scheduling problem to minimize resource leveling and makespan criteria

this paper considers a multi-skilled project scheduling problem that is newly developed form of Resource-constraint project scheduling problem (RCPSP). The main margin in such problems, compared with the classical RCPSP, is related to the given resource type. In fact, resources in MRCPSP are human with different skills. Each resource can only use one of its skills in a time unit. To do each activity in MPCPSP, different skills will be required. Additionally, the problem has two objective functions, in which the first one namely makespan is to minimize the project completion time. The second one, however, tries to minimize the total deviation of resource usage called resource leveling. Due to the complexity, two eminent meta-heuristic algorithms called NSGA-II and SPEA-II are employed. To represent the performance of the algorithms, a set of problem with different size are applied to the algorithms. Finally, the efficiency of the algorithms is compared with different methods.

Keywords:

Multi-skill project scheduling , Multi-objective optimization , Resource leveling , SPEA-II , NSGA-II , makespan