Task package division method for the integrated scheduling framework of mixed model car-sequencing problem. (July 2022)

• Record Type: Journal Article
• Authors: Wang, Hui; Sheng, Buyun; Lu, Xincheng; Fu, Gaocai; Luo, Ruiping
• Is Part Of: Computers & industrial engineering. Volume 169(2022)
• Publication Date: 2022-07
• Subjects: Mixed model car-sequencing problem -- Integrated scheduling framework -- Task package division -- Evolutionary algorithm -- Solution information; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2022.108144
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Propose an integrated scheduling framework of mixed model car-sequencing problem. Model a task package division problem considering inventory and scheduling factors. Designs an evolutionary algorithm based on solution information. Provide trade-off between sequence transformation and workshops performance. Develop a scheduling platform based on the integrated scheduling framework. Abstract: Car mixed model production can realize that cars with different models can be produced in a same production line at the same time, which is an effective way to respond the diversified market demand timely. How to design the optimal and feasible car sequences for multi car workshops is a key point for each car manufacturing enterprise. This paper presents an integrated scheduling framework of mixed model car-sequencing problem that includes order sequencing, task package division, and integrated scheduling of mixed model workshops. In this paper, task package division problem as the middle link of the integrated scheduling framework is concerned and investigated, considering the stereo garbage inventory and subsequent scheduling factors of multi workshops. Then, a mathematic model is constructed optimizing three objectives: (1) maximize the proportion of extra inventory cars in buffer; (2) minimize the number of car type changeover and color changeover in the task package; (3) minimize the difference of car type and quantity in the task package. Furthermore, an evolutionary algorithm based on solution information is presented to solve the model. In the proposed algorithm, three strategies adopting domain knowledge and solution information generated on iteration process are utilized to improve the searching ability and efficiency of the algorithm. The validity of the algorithm is verified by a real case. The computational results reveal that the proposed model and algorithm obtain appropriate results and have the potential to be applied to other similar problems.