An integrated model for strategic supply chain design: Formulation and ABC-based solution approach

• Published in: Expert systems with applications 2016-06, Vol.52, p.39-49
• Main Authors: Zhang, Linda L., Lee, Carman, Zhang, Shuzhu
• Format: Article
• Language: English
• Subjects: ABC algorithm; Bill of materials; Business administration; Design engineering; Expert systems; Humanities and Social Sciences; Mathematical models; Mixed integer; Multi-product; Multi-stage; Nonlinear programming; Procurement; Strategic supply chain design; Supplier priority; Supply chains; Swarm intelligence
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2015.12.035

•It presents an integrated model for supply chain design.•It considers three stages and their interactions in the model.•It develops an ABC-based solution approach.•It demonstrates the efficiency of the ABC-based approach. This study develops a mixed integer nonlinear programming (MINLP) model to design supply chains. In view of the limitations of many available strategic supply chain design models, this model involves three major supply chain stages, including procurement, production, and distribution, and their interactions; it takes into account bill of materials constraints for modeling complex supply chain inter-relationships. In addition, in accordance with the fact that companies nowadays develop product families, our model addresses multi-product supply chain design to respond to diverse customer requirements. Recognizing their importance, this study identifies and formulates constraints related to facility pairwise relationships and supplier priority along with the classical constraints from the available literature. To efficiently solve such a highly constrained, large scale MINLP model, we develop an approach based on an artificial bee colony (ABC) algorithm. Bicycle design and production is used to demonstrate the potential of the MINLP model for designing supply chains and the performance of the ABC-based solution approach in solving the model. The proposed model and solution approach can be considered as two fundamental components of an expert system in the broad sense. Thus, this study is expected to stimulate more future research on the development of practical expert systems for designing supply chains.