Multi-objective sustainable process plan generation for RMS: NSGA-III vs New NSGA-III
Résumé
Nowadays, to be relevant, the manufacturing system of a company has to be simultaneously cost and time-efficient and environmentally harmless. The RMS paradigm is proposed to cope with these new challenges. This paper addresses a multi-objective sustainable process plan generation problem in a reconfigurable manufacturing context. A non-linear multi-objective integer program (NL-MOIP) is proposed, where four objectives are minimized: the amount of greenhouse gas emitted by machines, the hazardous liquid wastes, the classical total production cost, and the total production time. To solve the problem, adapted versions of the well-known non-dominated sorting genetic algorithm (NSGA) approach, namely NSGA-III and New NSGA-III, are developed. Finally, the evaluation of the efficiency of the two approaches is performed through the use of four metrics: cardinality of the Pareto front (CPF), the cardinality of the mixed Pareto fronts (CMPF), inverted generational distance (IGD), and diversity metric (DM).