Multi-criteria supplier selection problem with fuzzy demand: a newsvendor model

In sourcing decisions, a buyer encounters three main issues. The first issue is the demand uncertainty that leads the buyer to find the optimal inventory level that balances inventory costs and customer satisfaction. The second one is deciding which suppliers to buy from and how much to order from each. Suppliers may have limited supply capacity that causes the buyer to split his/her order among multiple suppliers. To determine the suppliers’ order quantities, the buyer should evaluate them on the basis of factors such as wholesale price and quality. The third issue is the buyer’s purchasing strategy that can be reflected by assigning various weights to these factors. To help the buyer deal with these three issues, we propose a two-stage solution approach. In Stage 1 the Technique for Order of Preference by Similarity to Ideal Solution is employed to evaluate and rank the suppliers based on multiple factors. In Stage 2 a newsvendor problem is formulated where the uncertain demand is described by a fuzzy number having a general membership function. Then, an algorithm is employed to solve the newsvendor problem and assign the order quantities to the sorted suppliers to optimize the buyer’s inventory level. We use a numerical analysis to demonstrate the accuracy and effectiveness of the proposed solution approach. The numerical analysis also compares different types of uncertain demand in optimizing the buyer’s inventory level.