A re-entrant hybrid flow shop scheduling problem with machine eligibility constraints
研究来自真实转子车间的生产调度问题，建立混合整数线性规划模型并用Cplex求解，提出带改进交叉算子的离散差分进化算法，通过测试验证了算法和新型解码器的有效性。
A production scheduling problem originating from a real rotor workshop is addressed in the paper. Given its specific characteristics, the problem is formulated as a re-entrant hybrid flow shop scheduling problem with machine eligibility constraints. A mixed integer linear programming model of the problem is provided and solved by the Cplex solver. In order to solve larger sized problems, a discrete differential evolution (DDE) algorithm with a modified crossover operator is proposed. More importantly, a new decoder addressing the machine eligibility constraints is developed and embedded to the algorithm. To validate the performance of the proposed DDE algorithm, various test problems are examined. The efficiency of the proposed algorithm is compared with two other algorithms modified from the existing ones in the literatures. A one-way ANOVA analysis and a sensitivity analysis are applied to intensify the superiority of the new decoder. Tightness of due dates and different levels of scarcity of machines subject to machine eligibility restrictions are discussed in the sensitivity analysis. The results indicate the pre-eminence of the new decoder and the proposed DDE algorithm.