MGREVB-RNWD data

Due to the pressure of resource shortage and environmental protection, recycling retired electric vehicle batteries (REVBs) becomes a critical problem of many enterprises, such as electric vehicle (EV) manufacturers, EV battery manufacturers and third-party recycling service providers (3PRs), especially when the quantity of REVBs is uncertain. To recycle REVBs efficiently under uncertainty, this article mainly focuses on a multi-period green REVB recycling network (REVB-RN) design problem with winner determination (WD), denoted by MGREVB-RNWD, from the perspective of a fourth-party logistics provider (4PL). The novelty of MGREVB-RNWD comes from two aspects. First, to address quantity uncertainty of REVBs, WD is initially integrated with REVB-RN to select suitable 3PRs via combinatorial reverse auctions. Second, to improve sustainability of REVB-RN, the policies of carbon emissions cap and carbon tax are included. To minimize the total cost of recycling REVBs, a two-stage stochastic MGREVB-RNWD model is developed, which can be transformed to an equivalent mixed integer programming model by introducing random samples and auxiliary variables. To address the difficulty of introducing a huge number of samples, a sample-based decomposition algorithm (SBDA) is developed for solution method according to Latin hypercube sampling, sample average approximation and progressive hedging approaches. Numerical experiments are conducted to show the effectiveness of the proposed model and algorithm by using a real case study. Managerial implications are provided to design a economical and sustainable REVB-RN, where the adoption of WD to select suitable 3PRs can reduce the total cost, improve the recycling efficiency and lower the carbon emissions.