https://hal-utt.archives-ouvertes.fr/hal-02494741Belkaid, FayçalFayçalBelkaidMELT - Manufacturing engineering laboratory of Tlemcen - Université Aboubekr Belkaid - University of Belkaïd Abou Bekr [Tlemcen]Yalaoui, FaroukFaroukYalaouiLOSI - Laboratoire d'Optimisation des Systèmes Industriels - ICD - Institut Charles Delaunay - UTT - Université de Technologie de Troyes - CNRS - Centre National de la Recherche ScientifiqueSari, ZakiZakiSariMELT - Manufacturing engineering laboratory of Tlemcen - Université Aboubekr Belkaid - University of Belkaïd Abou Bekr [Tlemcen]Investigations on Performance Evaluation of Scheduling Heuristics and Metaheuristics in a Parallel Machine EnvironmentHAL CCSD2016Genetic AlgorithmSchedule ProblemParallel MachineMixed Integer Linear ProgrammingHybrid Genetic Algorithm[INFO.INFO-RO] Computer Science [cs]/Operations Research [cs.RO][INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS]Gavrysiak, Daniel2020-02-29 10:58:182023-02-08 17:11:112020-02-29 10:58:18enBook sections10.1007/978-3-319-23350-5_91Scheduling problems with consumables resources are common in many operations management and typically in industrial production practice. However, a significant part of scheduling problems studies deal with resources that are always available, but this assumption cannot be satisfied in many practical situations. This paper presents the results of a simulation study of parallel machines environment when each job is characterized by different non-renewable resources requirements. Each resource is delivered at different times following a cumulated arrivals stairs curves. The efficiency measure is the makespan . To describe the problem more clearly, a mathematical programming model is presented. This model represents a realistic and complex situation, in which jobs affectation, sequencing and resource assignment decisions are considered simultaneously. Due to its complexity, we decided to address this problem by means of a metaheuristic based genetic algorithm. Subsequently an improvement phase dealing with a local search method is proposed to improve the efficiency of the algorithm. Moreover, some heuristics are developed to deal with this problem. A simulation study is carried out on a set of test instances. The results are compared on the basis of computational time and solution quality. The simulations show that the hybrid genetic algorithm is able to find an optimal solution for small-sized problems within a reasonable computation time; also it outperforms genetic algorithm and heuristics methods for large-sized problems. These results validate the efficiency of the proposed algorithm.