Batch flow hybrid flow shop scheduling method based on adaptive evolutionary algorithm

By combining the three-layer encoding of the adaptive evolutionary algorithm with the self-evolution of the grouped population to improve the N7 neighborhood structure, the problems of low search efficiency and poor solution quality in the batch flow mixed flow workshop scheduling problem are solved, and efficient allocation of computing resources and optimization effect are achieved.

CN120875339APending Publication Date: 2025-10-31JIANGHAN UNIVERSITY
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202510931617.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-07
Publication Date
2025-10-31

AI Technical Summary

Technical Problem

Existing technologies suffer from low search efficiency and poor solution quality when solving batch-flow mixed flow shop scheduling problems. They also make it difficult to achieve effective allocation and rational utilization of computing resources. In particular, they lack specificity when dealing with scheduling schemes with different batch partitioning characteristics, which affects the convergence and solution quality of the algorithm.

Method used

A batch-flow hybrid pipeline scheduling method based on adaptive evolution algorithm is adopted. Individuals are encoded through a three-layer coding structure, and self-evolution is performed in grouped populations. Combined with an improved N7 neighborhood structure, local search is performed, and the population structure is dynamically adjusted to realize differentiated evolution strategies and adaptive allocation of computing resources.

Benefits of technology

It significantly improves the convergence speed and optimization quality of the algorithm, achieves full search under the excellent batch partitioning method and saves computational resources under the suboptimal batch partitioning method, and improves the solution efficiency and quality of the algorithm.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120875339A_ABST
    Figure CN120875339A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of workshop production scheduling, in particular to a batch flow hybrid flow workshop scheduling method based on an adaptive evolutionary algorithm, which comprises the following steps of: coding individuals by adopting a three-layer coding structure; generating an initial population and putting individuals in the initial population into different feature groups according to different batch division features; individuals in each group are self-evolved, high-quality individuals are locally searched, and the current optimal individual is recorded; self-adaptively adjusting the number of individuals in each group through the overall evaluation of each group; eliminating inferior individuals in each group according to the self-adaptive probability; and outputting an optimal individual when a termination condition is reached. According to the method, the hybrid evolutionary algorithm with a batch division adaptive adjustment mechanism is constructed, through adaptive adjustment of the number of individuals in each batch feature group in the evolutionary process, full mining is carried out in a high-quality batch division mode, computing resources are saved in a suboptimal batch division mode, and the efficiency is improved. And the search efficiency and the solving quality of the algorithm can be improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Flexible job shop multi-target scheduling method and system based on improved genetic algorithm

    CN113822525A