A multiobjective simulation optimization approach to define teams of workers in stochastic production systems

In this paper, we address team configuration problems in manufacturing systems, which consist in defining the number of workers to be assigned to a production system, as well as the skills that each worker must have in order to meet several performance measures. This problem is studied in a stochastic production context. A multi-objective evolutionary algorithm is connected to a simulation model to deal with this problem. Two objectives are considered. The first one is the minimization of the expected manpower cost associated to manufacturing team and the second one is the minimization of the expected mean flow time of jobs. Machines redundancy and workers multi-functionality are considered, when defining workers skills, to cope with possible random events such as workers unavailability and bottlenecks. Since the way workers are assigned to work centers strongly impact the results, a recent adaptive assignment heuristic is embedded in the simulation model and its parameters are also optimized. The proposed multi-objective simulation optimization approach is applied to design manufacturing teams, of a job shop production system, using the Nondominated Sorting Genetic Algorithm II (NSGA-II) connected to a simulation model developed using Arena. The set of non dominated solutions is found, so that an additional multi-criteria analysis can be performed.