5. Discussion and conclusions
Supplier selection and order allocation are critical problems in GSCM. Few prior studies have considered both problems simultaneously. The indicators or criteria for green supplier performance evaluation can be divided into qualitative and quantitative types. The indicator values obtained through measurement produce quantitative data, such as the defective rate, stock-out rate and price. Some indicators entail subjective assessment, for example, of environmental performance, service quality, reputation, information security, etc. Combining both types of indicators to measure supplier performance is a complex problem. In addition, the different viewpoints of the various department managers in the case company make the process of linking supplier performance with order allocation more difficult. For example, the purchasing department considers cost to be most important, while quality assurance deems quality to be the most essential factor. Therefore, this study combined MADM and MODM models to establish an integrated model to help companies select the most qualified suppliers and allocate orders to them. Our proposed model improved on the model proposed by Kannan et al. (2013) in several aspects. First, we obtained the criteria weights from BWM, which is an efficient and effective method of deriving the criteria weight vectors in the analyzed MADM problems because it requires fewer pairwise comparisons and easily obtains consistent results. Second, ranking indices in our model were calculated using modified TOPSIS, as proposed by Kuo (2017). Ranking indices determined using modified TOPSIS are superior to those attained using traditional TOPSIS when there are more than two alternatives, because modified TOPSIS considers the relative gaps of all alternatives and the weights of the distances from the PIS and NIS. Third, this study considered several objective functions (i.e., cost, delivery performance, product quality, and total utility) in constructing the MOLP model.
