Hybrid multi-objective opposite-learning evolutionary algorithm for integrated production and maintenance scheduling with energy consideration

While conventional scheduling researches take production efficiency, cost and quality as objectives, increasingly serious ecological problems and energy shortage have turned scholars’ attention to energy-efficient scheduling. Meanwhile, maintenance activities are of great importance to equipment ava...

Full description

Saved in:
Bibliographic Details
Published in:Neural computing & applications 2021-03, Vol.33 (5), p.1587-1605
Main Authors: Zhou, Binghai, Li, Xiujuan, Liu, Wenlong
Format: Article
Language:English
Subjects:
Artificial Intelligence
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data Mining and Knowledge Discovery
Energy consumption
Evolutionary algorithms
Genetic algorithms
Image Processing and Computer Vision
Machine learning
Maintenance management
Original Article
Preventive maintenance
Probability and Statistics in Computer Science
Production scheduling
Scheduling
Shutdowns
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:While conventional scheduling researches take production efficiency, cost and quality as objectives, increasingly serious ecological problems and energy shortage have turned scholars’ attention to energy-efficient scheduling. Meanwhile, maintenance activities are of great importance to equipment availability and production continuity. This paper addresses an energy-efficient multi-objective scheduling problem of a serial production line (SPSP) integrating production and maintenance, with the criteria of minimizing the makespan and the total energy consumption simultaneously. The impact of integrating preventive maintenance (PM) is analyzed, and the better interval mode is picked out. An energy-saving strategy combining shutdown windows of PM and energy-saving windows on idle machines is designed to cut down the total energy consumption. In order to tackle this NP-hard problem, a hybrid multi-objective opposite-based learning evolutionary algorithm (HMOLEA) is developed, in which the opposite-based learning and IGD indicator-based evolutionary mechanism are combined together and a special rank assignment is designed to improve the computational efficiency. Furthermore, a self-adaptive weighted mutation operator is fused into the framework of HMOLEA to enhance the exploration of the algorithm. Extensive computational experiments are carried out to verify the effectiveness and superiority of HMOLEA in solving the SPSP.
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-020-05075-3