A Comparative Study of Two Combinatorial Reverse Auction Models

Online group-buying is one of the most innovative business models employed by many companies. From the perspective of buyers, quantity based discounts provide a huge incentive to form coalitions and take advantage of lower prices without ordering more than their actual demand. Traditional group-buying mechanisms are usually based on a single item and uniform cost sharing. One way to reduce the cost for acquiring the required items is to take into account the complementarities between items provided by the sellers. By holding a combinatorial reverse auction, the total cost to acquire the required items will be significantly reduced due to complementarities between items. However, combinatorial reverse auctions suffer from high computational complexity. If there are multiple buyers, there are two different business models for procurement based on combinatorial reverse auctions: (1) independent combinatorial reverse auctions: each buyer may hold a combinatorial reverse auction independently and (2) combinatorial reverse auctions based on group buying: multiple buyers delegate the auction to a group buyer and the group buyer holds only one combinatorial reverse auction for all the buyers. In developing an effective tool to support the decision of multiple buyers’ procurement, a comparative study on the performance and efficiency of these two different business models is needed. In this paper, we compare the performance as well as the computational efficiency for these two combinatorial reverse auction models. Our analysis indicates that group buying combinatorial reverse auction outperforms multiple separate combinatorial reverse auctions not only in performance but also in efficiency.