The no-idle flowshop problem(NIFSP) is a typical NP-hard problem with constraints in the flowshop problem. In most real-world scenarios, the no-idle constraint is the basic requirement for the machine. However, the current proposed algorithms on the NIFSP problem have poor flexibility in comprehensive performance and parameter adjustment, for large-scale instances. To this end, this paper proposed an variable block internal iteration algorithm named VIIA with the goal of minimizing the maximum makespan. In the initialization phase of VIIA, it used the improved FRB5 produced the initial solutions, which improved the efficiency of FRB5, and greatly reduced the CPU consumption time while ensuring the algorithm performance. In the destruction reconstruction phase, it increased the internal iteration of the number of removed variable blocks, thus flexibly adjusted the parameter values. VIIA increased the neighborhood search to adapted to different size of instances. In order to verify the performance of VIIA, the algorithm was compared with several algorithms that perform well in the flowshop scheduling problem. The experimental results demonstrate the superiority of VIIA in solving the NIFSP problem, and the performance of VIIA is obviously better than that of the comparison algorithms in the search of the optimal solution.

[2022-08-02] Accepted Paper /doc/2022.05.0234/v1

[2022-12-05] Printed Article /doc/2022.05.0234/v2