A distributed blocking flow-shop scheduling optimizer driven by a learning mechanism

By employing a learning mechanism-driven distributed congested flow shop scheduling optimizer, which utilizes multi-group collaborative adaptive search and elite strategies, the total delay time and total energy consumption in the distributed congested flow shop scheduling problem are optimized. This achieves bidirectional optimization of production efficiency and energy consumption, thereby improving production effectiveness.

CN116203905BActive Publication Date: 2026-06-26LANZHOU UNIVERSITY OF TECHNOLOGY +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
LANZHOU UNIVERSITY OF TECHNOLOGY
Filing Date
2023-03-12
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing technologies are insufficient to effectively optimize total delay time and total energy consumption in the distributed congested flow workshop scheduling problem. Traditional mathematical methods cannot accurately establish mathematical models, and intelligent optimization algorithms lack the ability to explore and develop solutions for high-dimensional problems.

Method used

A learning-driven distributed congested pipeline workshop scheduler optimizer is designed by using multi-group cooperative adaptive evolutionary search, combining elitist strategies and historical information, and considering the total latency and total energy consumption. It performs global and local searches and performs acceleration on critical paths and deceleration on non-critical paths.

Benefits of technology

It achieved a comprehensive balance between production efficiency and energy consumption indicators, reduced the maximum total delay and total energy consumption, improved production efficiency, and enhanced population diversity and exploration capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116203905B_ABST
    Figure CN116203905B_ABST
Patent Text Reader

Abstract

The application provides a distributed blocking flow shop scheduling optimizer driven by a learning mechanism, and proposes a distributed estimation algorithm applying reverse learning and differential evolution to optimize energy-saving distributed blocking flow shop scheduling. The application fully considers the energy consumption in actual production, and designs an initialization method considering total delay time and total energy consumption; in order to improve the quality of the population, a multi-population collaborative operation guided by reinforcement learning and reverse learning is designed, information interaction is realized through the multi-population collaborative mode, and the search speed is accelerated. Based on the specific characteristics of different populations, adjustable parameter variables meeting the exploration and development capacity are designed. In order to optimize the distributed blocking flow shop scheduling problem with the objective function of reducing total delay time and total energy consumption, acceleration and deceleration operations on different paths are proposed. Through comparison on the 2017 test set and examples composed of different numbers of machines, workpieces and machine arrays, the overall performance of the optimizer is better than that of other optimizers.
Need to check novelty before this filing date? Find Prior Art