Sustainable supplier selection and order allocation: Distributionally robust goal programming model and tractable approximation. (February 2020)

• Record Type: Journal Article
• Authors: Jia, Ruru; Liu, Yankui; Bai, Xuejie
• Is Part Of: Computers & industrial engineering. Volume 140(2020)
• Publication Date: 2020-02
• Subjects: Supplier selection and order allocation -- Sustainability -- Ambiguity set -- Tractable approximation; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2020.106267
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: The imprecise distributions of uncertain input data are considered and characterized by ambiguity sets. A novel distributionally robust sustainable SS/OA goal programming model is firstly proposed. A new tractable safe approximation model is deduced under ambiguity sets. A case study is provided to demonstrate the effectiveness of the proposed model. Abstract: A challenge of planning real-life sustainable supplier selection and order allocation (SS/OA) problems for purchasing companies is to gather the extensive data and exact distributions of input data. Motivated by this challenge, this paper addresses imprecise probability distributions of uncertain per unit cost, CO 2 emissions, demand, supply capacity and the minimum order quality and characterizes these distribution uncertainty by ambiguity sets including the true distributions. Moreover, the purchasing company decisions to be optimized simultaneously include four conflicting objectives regarding cost, CO 2 emissions, society and suppliers' comprehensive value while considering risk measures incurred by cost and CO 2 emissions to achieve company sustainability. To help decision makers formulate practicable policy, a novel distributionally robust sustainable SS/OA goal programming model is developed with expected constraints and joint chance constraints. The proposed optimization model can balance multiple conflicting objectives, and effectively solve our sustainable SS/OA problem. More importantly, we structure ambiguous distributions sets, and thus derive the computationally tractable approximation form of the proposed practical model. Finally, we illustrate our optimization method through a case study about a steel company, conduct a thorough inquiry into the effect of uncertainty and summarize the management implications of the results.