Elsevier

Computers in Industry

Volume 56, Issue 2, February 2005, Pages 161-168
Computers in Industry

Genetic algorithms for integrated preventive maintenance planning and production scheduling for a single machine

https://doi.org/10.1016/j.compind.2004.06.005Get rights and content

Abstract

Despite the inter-dependent relationship between them, production scheduling and preventive maintenance planning decisions are generally analyzed and executed independently in real manufacturing systems. This practice is also found in the majority of the studies found in the relevant literature. In this paper, heuristics based on genetic algorithms are developed to solve an integrated optimization model for production scheduling and preventive maintenance planning. The numerical results on several problem sizes indicate that the proposed genetic algorithms are very efficient for optimizing the integrated problem.

Introduction

Production scheduling and preventive maintenance (PM) planning are among the most common and significant problems faced by the manufacturing industry. Production schedules are often interrupted by equipment failures, which could be prevented by proper preventive maintenance. However, recommended PM intervals are often delayed in order to expedite production. Despite the trade-offs between the two activities, they are typically planned and executed independently in real manufacturing settings even if manufacturing productivity can be improved by optimizing both production scheduling and PM planning decisions simultaneously.

Numerous studies have been conducted in these two areas in the past decades. Shapiro [1] and Pinedo [2] reviewed various papers in production scheduling. Similarly, Sherif and Smith [3] and Dekker [4] reviewed several studies using maintenance optimization models. However, almost all relevant studies considered production scheduling and PM planning as two independent problems and therefore solve them separately.

Only a few studies have tried to combine and solve both problems simultaneously. Graves and Lee [5] presented a single-machine scheduling problem with the objective to minimize the total weighted completion time of jobs. However, only one maintenance activity can be performed during the planning horizon. Lee and Chen [6] extended Graves and Lee's research to parallel machines, but still permitting only one maintenance action. Qi et al. [7] considered a similar single-machine problem with the possibility for multiple maintenance actions, but the risk of not performing maintenance is not explicitly included in the model. Cassady and Kutanoglu [8] developed an integrated mathematical model for a single-machine problem with total weighted expected completion time as the objective function. Their model allows multiple maintenance activities and explicitly captures the risk of not performing maintenance.

In this paper, we develop genetic algorithm heuristics to solve the integrated production scheduling and preventive maintenance planning problem for a single machine introduced in Cassady and Kutanoglu [8]. The following section, Section 2, contains an overview of the integrated production scheduling and PM planning problem. Section 3 briefly describes the proposed genetic algorithm procedures. The experimental results of multiple problem sizes appear in Section 4. The conclusions are summarized in Section 5.

Section snippets

Integrated production scheduling and PM planning problem

This section describes the integrated model and proposed solution procedures for a single-machine production scheduling and PM planning problem presented by Cassady and Kutanoglu [8].

Genetic algorithm procedures

Initially, genetic algorithm (GA) procedures were mainly applied to research topics in the area of artificial intelligence. However during the past decade, GA has become one of the most well-known search heuristics and is widely used in many combinatorial optimization problems including machine scheduling. Chambers [9] reviewed applications of GA in several research areas. For the single-machine scheduling problem, Gupta et al. [10] applied GA to minimize the variance of flow time. Liu and Tang

Results

In this section, the proposed GA-based heuristics (GA1, GA2, and GA3) are compared with the total enumeration approach (TE) and the heuristic (H) suggested by Cassady and Kutanoglu [8]. Three problem sizes (small, medium, and large) are used to compare the performance of the algorithms. All algorithms are evaluated for the small size problems. Because of the excessive computation of the enumerative strategies, TE is only considered for the small size problems and H is not considered for the

Conclusions

In this study, the genetic algorithm procedure is successfully applied to the integrated optimization model for production scheduling and preventive maintenance planning proposed by Cassady and Kutanoglu [8]. Three GA-based algorithms are developed. Their performance is evaluated using multiple instances of small, medium, and large size problems. Based on the results, we conclude that the proposed genetic algorithms can be used to effectively solve the integrated problem. Future work includes

Acknowledgement

The authors thank the anonymous referees and the members of the Editorial Board for the comments and insights. Their feedback led to significant improvements to the content and presentation of this work.

Navadon Sortrakul is an optimization engineer team leader at Transplace, Inc. Before his current position, he has worked at J.B. Hunt Logistics, Inc. He received his PhD in Industrial Engineering from the University of Arkansas, his MS in Industrial Engineering from the Iowa State University, and his BS in Industrial Engineering from the Chulalongkorn University, Thailand. His primary research interests are in applied operations research in production scheduling, preventive maintenance

References (14)

There are more references available in the full text version of this article.

Cited by (136)

  • Optimal mission aborting in multistate systems with storage

    2022, Reliability Engineering and System Safety
View all citing articles on Scopus

Navadon Sortrakul is an optimization engineer team leader at Transplace, Inc. Before his current position, he has worked at J.B. Hunt Logistics, Inc. He received his PhD in Industrial Engineering from the University of Arkansas, his MS in Industrial Engineering from the Iowa State University, and his BS in Industrial Engineering from the Chulalongkorn University, Thailand. His primary research interests are in applied operations research in production scheduling, preventive maintenance optimization, and logistics. He is a member of IIE, INFORMS, and Phi Kappa Phi.

Heather Nachtmann is an assistant professor of Industrial Engineering at the University of Arkansas. She joined the UofA faculty in 2000. She received her PhD in Industrial Engineering from the University of Pittsburgh. Her research interests include economic decision analysis, engineering valuation, and applied operations research. Her primary research applications are intermodal transportation networks and logistic systems. She holds leadership positions in ASEE and IIE. She is a member of AACE International, ASEM, and INFORMS.

C. Richard Cassady is an associate professor in the Department of Industrial Engineering at the University of Arkansas. Prior to joining the faculty at the UofA, he served on the industrial engineering faculty at Mississippi State University. He received his BS summa cum laude, MS and PhD, all in industrial and systems engineering, from Virginia Tech. His primary research interests are in repairable systems modeling, including the evaluation and optimization of equipment maintenance policies. He also conducts research in reliability engineering, statistical quality control and applied operations research. He is a Senior Member of IIE and a Member of ASEE, ASQ, INFORMS and SRE. He is also a member of the RAMS Management Committee.

View full text