Hybrid simulated annealing and MIP-based heuristics for stochastic lot-sizing and scheduling problem in capacitated multi-stage production system

详细信息    查看全文

• 作者：Reza Ramezanian ; ^{Ramezanian@iust.ac.ir} ; Mohammad Saidi-Mehrabad ^{Mehrabad@iust.ac.ir}
• 关键词：Stochastic lot-sizing and scheduling ; Flow shop ; Sequence-dependent setup ; MIP-based heuristic ; Hybrid meta-heuristic ; Taguchi method
• 刊名：Applied Mathematical Modelling
• 出版年：2013
• 出版时间：1 April, 2013
• 年：2013
• 卷：37
• 期：7
• 页码：5134-5147
• 全文大小：724 K

文摘

This paper addresses lot sizing and scheduling problem of a flow shop system with capacity constraints, sequence-dependent setups, uncertain processing times and uncertain multi-product and multi-period demand. The evolution of the uncertain parameters is modeled by means of probability distributions and chance-constrained programming (CCP) theory. A new mixed-integer programming (MIP) model with big bucket time approach is proposed to formulate the problem. Due to the complexity of problem, two MIP-based heuristics with rolling horizon framework named non-permutation heuristic (NPH) and permutation heuristic (PH) have been performed to solve this model. Also, a hybrid meta-heuristic based on a combination of simulated annealing, firefly algorithm and proposed heuristic for scheduling is developed to solve the problem. Additionally, Taguchi method is conducted to calibrate the parameters of the meta-heuristic and select the optimal levels of the algorithm¡¯s performance influential factors. Computational results on a set of randomly generated instances show the efficiency of the hybrid meta-heuristic against exact solution algorithm and heuristics.