A multi-attribute double auction and bargaining model for emergency material procurement

Abstract

Emergency material procurement (EMP) is not only a significance segment of the emergency preparedness, but also an important guarantee for disaster relief. The emergency materials have the different quality attributes towards ordinary commodities, so attribute requirements need to be considered in the procurement activity. This research designed a two-stage MADA-B mechanism to study the supply-demand matching of emergency materials with multiple attributes. In the first stage, considering the differentiated needs of multiple buyers (the government and non-governmental organizations) and the differentiated capacity of multiple suppliers, a multi-attribute double auction (MADA) model was constructed. We analyzed the benchmark bidding strategy of buyers and sellers, and then proposed a winner determination model whose objective is maximizing the total bid-ask spreads. In the second stage, considering the buyer's and the seller's discount rate as private information, an infinite bargaining model was established. We analyzed the bargaining process and gave the method to determine the transaction price and quantity. After that, we designed a numerical example to simulate the practice under the background of purchasing multi-attribute emergency ration package (ERP). The research shows that the auction-bargaining mechanism can not only match the attribute between buyers and sellers, but also can scientifically determine the transaction price and quantity according to the wishes of both sides. It has good feasibility and applicability in the emergency management practice. • A multi-attribute double auction and bargaining mechanism is proposed. • It aims to solve the participant matching problem in emergency material procurement. • It can reach an equilibrium on attribute matching, transaction price and quantity. • The strategies of participants in both auction and bargaining stages are given. • The bargaining has no period restriction and is under incomplete information.