Research On No-Wait Flow Shop Scheduling Problem Considering Learning Effect

No-wait flow shop scheduling problem is an important branch of flow shop scheduling problem.The constraint of no-wait is closer to the processing links of iron and steel making,biological pharmaceutical,chemical processing and others.In this paper,based on the processing of mill rolls,two different no-wait flow shop scheduling models with the position-based learning effect are considered.The main tasks of this paper are as follows.(1)We focus on a no-wait permutation flow shop scheduling problem considering learning effect with the goal of minimizing the makespan.The learning effect is based on the truncation function of the machining position of the job,in which the processing time of the job decreases to a certain extent and tends to be stable.In order to solve this problem,a genetic-path relinking search algorithm is designed.Based on the genetic algorithm,elite retention strategy,path relinking operator and local search operator are added to strengthen the optimization effect.The computational experiment show that the hybrid algorithm is significantly better than the contrast algorithms in terms of convergence speed,optimal solution quality and robustness.(2)We investigate a no-wait hybrid flow shop scheduling problem with the truncated learning effect for minimizing the makespan.The decision variables include the machine allocation of the job,the processing sequence and the starting processing time of the job in the first process.In order to solve the proposed model,an adaptive genetic-symbiotic organisms algorithm is proposed.The improved algorithm combines the adaptive genetic algorithm with the addition of mutualism operator,commensalism operator and parasitism operator based on roulette.The computational experiments show that the improved algorithm is superior to the contrast algorithms in terms of stability and ability to seek optimization.Considering the no-wait flow shop scheduling problem with learning effect and designing more efficient algorithms to help companies increase productivity have profound theoretical research value and practical application value.