A material handling scheduling method for mixed-model automotive assembly lines based on an improved static kitting strategy

https://doi.org/10.1016/j.cie.2020.106268Get rights and content

Highlights

  • Introduce an improved static kitting strategy for MMALs.

  • Apply the graph theory to describe the problem and propose a Kuhn-Munkres algorithm.

  • Propose an EOADDE algorithm to solve the scheduling problem.

Abstract

Since the diversification of customer demands poses a great challenge for manufacturing enterprises and the scheduling problem of material handling affects the efficiency of assembly lines, this paper proposes a novel scheduling method, an improved static kitting strategy, to solve the scheduling problems of the material handling for automotive mixed-model assembly lines (MMALs) based on line-integrated supermarkets. Firstly, an integer programming mathematical model is established with the objective of minimizing the number of logistic workers. Then, an improved static kitting strategy is presented to solve the problem and a model based on graph theories is constructed to transform the scheduling problem to a mathematical one. Afterwards, a Kuhn-Munkres algorithm and an elite opposition-based learning adaptive dynamic differential evolution algorithm, named EOADDE algorithm, is developed to solve the scheduling problem. The elite opposition-based learning (EOL) and self-adaptive operators are applied to the proposed EOADDE algorithm to enhance the local search ability and the convergence speed. Finally, computational experiments of the proposed algorithm are carried out compared with benchmark algorithms, and the feasibility and effectiveness of proposed methods are verified by results.

Introduction

Under the comprehensive influence of energy-saving, environmental protection requirements and highly-diversified consumer demands in the past few decades, the automobile industry has been facing serious challenges. Improving productivity and reducing production costs have become important issues for enterprises. Consequently, mixed-model assembly lines (MMALs) have been widely employed by automobile manufacturers to satisfy customers’ diversified demands (Jainury, Ramli, Rahman, & Omar, 2014). However, the scheduling problem of material handling is a complicated mission. In many automobile manufacturers, the handling cost of in-house parts accounts for 15–30% of the total production cost. The operating cost of the manufacturing systems can be greatly reduced by optimizing the strategy of material handling (Kilic & Durmusoglu, 2015). Therefore, it is of great theoretical significance and practical value to investigate the scheduling problem of material handling for MMALs.

In recent years, the scheduling problem of material handling for MMALs has attracted the attention of quite some researchers. Johansson and Johansson (2006) classified a strategy of material handling into batch supply, continuous supply and kitting. Battini, Faccio, Persona, and Sgarbossa (2009) believed that the material handling strategy can be divided into pallets to workstations, trolleys to workstations and kits to assembly lines. Subsequently, by summarizing the previous studies, Caputo and Pelagagge (2011) divided into line storage, kitting and just-in-time kanban-based continuous supply strategies. Now, most of the literatures refer to two main delivery strategies, namely, line stocking and kitting (Hua & Johnson, 2008). In the former concept, a line stocking strategy, also named as continuous supply or bulk feeding (Limère, Van Landeghem, & Goetschalckx, 2015), indicates that homogenous parts are stored directly in large containers at assembly stations. Large containers are replenished by forklifts, trolleys etc. from a material warehouse. In this case, the major advantages are that there are free of double-handling operations and much flexibility in case of unexpected events (Bozer & McGinnis, 1992). When there is an alteration of material demands or an occurrence of a defective part, assembly workers can quickly withdraw spare parts. However, these material containers occupy the limited space of stations at the line (Limère, 2012). Also, assembly workers always need to retrieve and select required parts from multiple containers when assembling several models, which will aggravate the workload of workers and increase the assembly time. On the other hand, the kitting strategy can be classified into two types, namely the traveling kitting and the static kitting strategies. In the traveling kitting strategy, kits are delivered to the first station and move together with assembling products (Caputo, Pelagagge, & Salini, 2015). The static kitting strategy means that materials are sorted into small kits at the decentralized logistic areas which are loaded in the wagons of vehicles and delivered to each station. Thus, the kitting system can quickly respond to flexible demands in the assembly process compared to the line stocking strategy to fulfill the diversity of models to be assembled. Also, using small kits can reduce the line-side inventory in all stations. Nevertheless, Faccio (2014) pointed out that each part of kits is managed through an initial picking activity, which increases the handling costs, and picking a defective part or failure in preparation of kits will significantly affect the assembly process. Moreover, a new concept named the line-integrated supermarket was introduced by Boysen and Emde (2014) and successfully employed in a North-American plant of a German automobile enterprise.

In this paper, a novel concept of the improved static kitting strategy is presented which unifies the advantages of kitting and line stocking strategies by applying the layout of line-integrated supermarkets. It integrates decentralized supermarkets of the kitting strategy into the assembly line and materials are supplied in bulk to every station by logistic workers, so called the improved static kitting strategy. Compared to traditional line stocking and kitting strategies, the strategy has the advantages of stable supply, saving labors and less inventory at the line. In the line-integrated supermarket, the storage area for materials to be assembled by the respective assembly workers is located directly next to the assembly line. Logistic containers named JIS-bins (Just-in-sequence bins) are used to accommodate required parts and to deliver them from the logistic area of a station to the moving conveyor. From the JIS-bin, an assembly worker withdraws the part required by the car having currently entered the station and assembles it during the production cycle. The JIS-bin is prepared by a group of logistic workers, who pick the parts according to the predefined production sequence and pack the picked parts into the JIS-bin. In this case, it is clear that the improved static kitting strategy makes do without vehicles to deliver kits to stations, so that it saves labors and decreases the amount of double handling compared to the traditional kitting strategy. Moreover, in case of unforeseen events, spare parts are readily available at stations and the effort to replace the defective parts is reduced to a minimum. These properties make the improved static kitting strategy an attractive managerial concept for in-house material handling not only in the automotive industry but also in many other high-variety circumstances.

The proposed novel delivery system of the improved static kitting strategy gives rise to some important decision problems to be solved. This paper focuses on the scheduling problem of the refilling of JIS-bins on the assembly line with the improved static kitting strategy. Given the number of stations and their deterministic material demands over time, a schedule for the refilling of JIS-bins should be completed to ensure that there are no stock-outs and the number of logistic workers should be minimized on the premise of a certain number of stations.

To cope with the scheduling problem of the material handling for MMALs, this paper employs two methods, the exact approach and the meta-heuristic algorithm. The contributions of this paper are summarized as the following three points.

  • A novel material handling scheduling method for MMALs based on the improved static kitting strategy in the layout of a line-integrated supermarket with the handling resource consideration is proposed in this paper. The objective of this problem is to minimize the total number of logistic workers over the planning horizon with the constraints of handling resources of logistic workers and to meet the operational performance requirement for the batch supply.

  • The graph theory is introduced to illustrate the scheduling problem and a Kuhn-Munkres algorithm is proposed to obtain exact solutions.

  • A meta-heuristic algorithm, namely EOADDE algorithm with an elite opposition-based learning (EOL) and self-adaptive operators is adopted to solve the scheduling problem and its effectiveness and competitiveness is proved by comparison with other two algorithms.

The remainder of this paper is organized as follows. Section 2 provides a brief literature review. In Section 3, the problem description and model formalization are provided for the material handling of automotive MMALs with detailed assumptions and notations. Section 4 discusses the bipartite graph model of the scheduling problem, presents a Kuhn-Munkres algorithm to solve the optimal matching problem and illustrates an example to detail the steps of the algorithm. The EOADDE algorithm is introduced in Section 5. Computational experiments are carried out to validate the performance of the proposed algorithm in Section 6. Finally, Section 7 draws conclusions and provides the prospect of future research.

Section snippets

Literature review

In the area of in-house logistics, the concepts of the line stocking and kitting strategies are extensively discussed and applied for material handling in the automobile industries. Due to the concept of line-integrated supermarkets unifying the advantages of the line stocking and kitting strategies, a brief review of the two strategies is necessary. Many researchers have made different kinds of studies on the comparison between line stocking and kitting strategies.

Bozer and McGinnis (1992)

Problem description

Fig. 1 schematically depicts the concept of a line-integrated supermarket. In a line-integrated supermarket, the storage area for parts to be assembled by the respective assembly workers is located directly next to the assembly line. Logistic containers named JIS-bins are used to accommodate required parts and to deliver them from the logistic area of a station to the moving conveyor. From the JIS-bin, an assembly worker withdraws the part required by the car having currently entered the

Bipartite graph model of scheduling problem

A bipartite graph is a special model in graph theories, which has significant applications in some assigning and scheduling problems, and in many cases, it can obtain an algorithm for solving out accurate and efficient optimal solutions. According to the properties and concepts of bipartite graph, this paper will transfer the scheduling problem of material handling to the problem of finding a maximum matching in a bipartite graph. The specific formal expression is listed as follows: Given the

The proposed EOADDE algorithm

In the last section, the KM algorithm was proposed to solve small-scale problems. It will take considerably long time to solve medium or large-scale problems. Consequently, we introduce a meta-heuristic algorithm which can randomly search optimal solution and get a suboptimal solution in a limited time. We use differential evolution algorithm (DE) as a main framework of an optimization algorithm. However, due to the complexity of the proposed problem in this paper, standard DE is difficult to

Computational experiments

When carrying out computational experiments, taking the number of logistic workers as an optimization performance index and considering the characteristics of MMALs, we select two main parameters to present the characteristics, factor A that is the number of assembly stations and factor B that is the number of production cycles. We implement our algorithms in Matlab2018a and solve them through a set of test instances on a PC with an Intel Core i7-8550U 1.8 GHz CPU and 8 GB RAM.

Conclusions

In this paper, we study the scheduling problem of the material handling for MMALs based on an improved static kitting strategy in the layout of the line-integrated supermarket. First, considering the application background of the line-integrated supermarket, we formally describe the scheduling problem of the material handling for MMALs, determine some hypotheses and establish an integer programming mathematical model. Subsequently, an improved static kitting strategy is introduced to solve the

CRediT authorship contribution statement

Binghai Zhou: Conceptualization, Validation, Writing - review & editing, Writing - review & editing, Project administration, Funding acquisition. Zhaoxu He: Methodology, Software, Formal analysis, Investigation, Data curation, Writing - original draft, Visualization.

Declaration of Competing Interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgement

This work was partly supported by the National Natural Science Foundation of China (Grant No.71471135)

Binghai Zhou was born in 1965, in Zhejiang Province, China. He received his Master degree in Industrial Engineering from Shanghai Jiaotong University in 1989 and Doctor degree in 1992, respectively. He is currently a professor in the Mechanical Engineering school of Tongji University, Shanghai, China. His research interests cover scheduling and simulation of discrete systems (manufacturing/logistics Systems), preventive maintenance modeling of equipment and artificial intelligent algorithm.

References (36)

Cited by (0)

Binghai Zhou was born in 1965, in Zhejiang Province, China. He received his Master degree in Industrial Engineering from Shanghai Jiaotong University in 1989 and Doctor degree in 1992, respectively. He is currently a professor in the Mechanical Engineering school of Tongji University, Shanghai, China. His research interests cover scheduling and simulation of discrete systems (manufacturing/logistics Systems), preventive maintenance modeling of equipment and artificial intelligent algorithm.

Zhaoxu He was born in 1997, in Hebei Province, China. He received his Bachelor degree in Industrial Engineering from Tongji University, Shanghai, China. Currently, he is a senior student. His current research interests include evolutionary algorithm, manufacturing system modeling and simulation.

View full text