Workshop production scheduling method and system based on genetic algorithm

Abstract

The invention discloses a workshop production scheduling method based on a genetic algorithm, and the method comprises the steps: constructing and training a decision tree model based on labeled orderdata, and inputting to-be-produced order data into the trained decision tree model to obtain the priority of a to-be-produced order; based on comprehensive consideration of a genetic algorithm, constraint conditions, encoding and decoding, fitness calculation and the like, finding an optimal production scheduling plan and finally, presenting a specified number of encoding schemes to production scheduling personnel in a graphical interface form for selection. Furthermore, the invention also discloses a workshop production scheduling system based on the genetic algorithm corresponding to the workshop production scheduling method. In the process of using the genetic algorithm to participate in decision making, the factors such as production efficiency, overall balance of the production line,multi-objective optimization and possible emergencies are comprehensively considered, the production efficiency is improved, the balance of the production line is realized, and the maximization of enterprise benefits is facilitated.

Description

technical field [0001] The invention belongs to the technical field of workshop production scheduling plan management, and in particular relates to a production scheduling method and system based on a genetic algorithm. Background technique [0002] At present, the scheduling in actual production mostly uses the static method of ERP (Enterprise Resource Planning) system, and the production scheduling plan is compiled according to the requirements of the order and the supply of material capacity. However, the actual production situation is often inconsistent with expectations. The ERP system cannot collect production data in real time, nor can it effectively schedule and configure production resources, resulting in waste of time and resources and increased production costs. [0003] In the research of some enterprises, we found that many enterprises have not yet paid attention to dynamic production scheduling. Enterprises often do not have strict requirements for production s...

AI Technical Summary

Problems solved by technology

[0004] At present, there are some research schemes that try to solve the problem of dynamic production scheduling, but these research results are often not suitable for the actual production, and have the following shortcomings: 1) The linearity of the production line is not considered, and the machines in actual production have a certain degree of correlation. The machines are connected to a production line. Once a product is placed on the production line, it must pass through all the machines on this line. However, most of the existing solutions only regard the machines as scattered point sets, instead of taking the production line as the basis. 2) The objective function basically only focuses on the minimization of order completion time, the goal is too single, and does not fully meet the actual production requirements; 3) Lack of consideration for the overall balance of the production line, lack of consideration for changing lines, different Practical factors such as product delivery urgency often focus on individual stations while ignoring the dynamic characteristics of the multi-variety production and process change of the final assembly line

In short, the existing production workshops of manufacturing enterprises mainly rely on manual scheduling, and the response and adjustment are not timely enough. Even if there are some developed intelligent scheduling solutions, they are not suitable for actual production.

Images