A framework for online reverse auction based on marketmaker learning with risk-averse buyer

One of the new approaches to purchasing and procuring goods and materials in the supply chain is the use of reverse auction. With the rapid and ever-expanding development of information technology and the Internet around the world, the use of Internet platforms for this type of procure has also been taken into account and has created online reverse auction method. In this paper, a new framework for the online reverse auction process is provided that takes both sides of the procurement process (buyer and seller). The proposed auction process is a multi-attribute semi-sealed multi-round online reverse auction. In this process, an online market-maker, with the prediction of the buyer scoring function, facilitates the seller's bidding process. To fit the function, a multi-layer perceptron neural network model is used. In this case, in addition to hiding the seller's scoring function, information is provided to sellers to improve the bid. Also, the methods of scoring by the buyer are defined Additive, Multiplicative and risk aversion, which is based on the theory of perspective. Within this framework, sellers improve their bids in each round using an optimization model. By simulating the auction process, the proposed framework was evaluated in comparison with an open auction, taking into account seller scoring criteria, seller profits, and number of auction rounds