Discount Auctions for Procuring Hetergeneous Items
نویسندگان
چکیده
Advent of business over Internet have given rise to a number of innovative trading mechanisms. In this work we propose a new auction mechanism, called as discount auctions, for procuring heterogeneous items. The buyer, who is the auctioneer, has an unit demand for M distinct items. The suppliers, who are the bidders, specify individual costs for each of the items. In addition, a supplier also specifies a discount function: a non-decreasing function over the number of items. This discount bid, in essence, conveys the individual costs for each of the items and the discount that can be availed based on the number of items bought. The winner determination problem faced by the buyer is to choose the optimal set of winning suppliers and their respective winning items such that the total cost of procurement is minimized. First we show that this problem is NP-hard upon reduction from the set covering problem. Next we propose two exact algorithms to solve the problem to optimality. The first one is a branch-and-bound algorithm, called as branch-on-supply (BoS), which does not use mathematical programming formulation but rather exploits the embedded network structure. The second is a suite of branch-and-cut algorithms. We derive valid inequalities to the integer programming formulation, which serve as cuts for the LP relaxation. A heuristic branching technique, called as branch-on-price (BoP), is proposed that branches on the current price of an item, which is partially supplied by more than one supplier. The design philosophies of the above are different in the sense that BoS searches for the optimal number of items from the suppliers, whereas BoP searches for the optimal price of the items. We compare the performance of these algorithms with extensive computational experiments. Key-words: e-procurement, auctions, integer programming, transportation problem, branch-and-bound, linear programming relaxation, duality, valid inequalities Enchère de type pondéré pour l’acquisition d’articles multiples Résumé : L’arrivée de l’Internet dans le monde des affaires a provoqué la naissance d’un certain nombre de mécanismes de ventes (relations client-fournisseur) innovants. Dans cette étude, nous proposons un nouveau mécanisme d’enchère appelé discount auctions pour l’acquisition d’articles multiples (de types et de tailles différents). Un acheteur (client dans notre cas) exprime une demande unitaire pour M différents articles. Un ensemble de vendeurs (fournisseurs dans notre cas) vont cherché à répondre à ces demandes. L’offre d’un vendeur est composée du prix de chacun des articles séparément et d’une fonction de pondération qui indique la remise offerte sur la base du nombre d’articles achetés (plus d’articles achetés et plus la remise sur le prix de vente est importante). Le problème rencontré par l’acheteur est d’identifier, parmi l’ensemble des tous les vendeurs, les vendeurs dits vainqueurs ainsi que les articles achetés chez chacun d’eux en minimisant le coût d’achat total. Dans un premier temps, nous avons prouvé que ce problème d’optimisation est NP-difficile en utilisant une réduction du problème dit de recouvrement. Nous avons développé deux algorithmes exacts de types banch-and-bound. Le premier, appelé branch-on-supply (BoS), n’utilise pas la formulation mathématique du problème mais exploite sa structure de type réseau. Le second est un ensemble d’algorithmes de types branch-and-cut. En effet, nous avons introduit des inégalités dites valides dans le programme linéaire en nombres entiers modélisant le problème pour servir comme des coupes pour la relaxation de type LP. Une heuristique de branchement, appelée branch-on-price (BoP), est proposée. Les philosophies de conception des deux algorithmes sont différentes. BoS cherche à déterminer le nombre optimal d’articles à acheter de chaque fournisseur. A l’inverse, BoP cherche à déterminer le prix optimal de chaque article. Pour comparer les performances de deux algorithmes, des expériences numériques ont été réalisées et analysées. Mots-clés : e-procurement, enchères, programmation en nombres entiers, problème de transport, méthode de séparation et évaluation, relaxation, dualité, inégalités valides
منابع مشابه
An Optimal Multi-Unit Combinatorial Procurement Auction with Single Minded Bidders
The current art in optimal combinatorial auctions is limited to handling the case of single units of multiple items, with each bidder bidding on exactly one bundle (single minded bidders). This paper extends the current art by proposing an optimal auction for procuring multiple units of multiple items when the bidders are single minded. The auction minimizes the cost of procurement while satisf...
متن کاملA Lotting Method for Electronic Reverse Auctions
An increasing number of commercial companies are using online reverse auctions for their sourcing activities. In reverse auctions, multiple suppliers bid for a contract from a buyer for selling goods and/or services. Usually, the buyer has to procure multiple items, which are typically divided into lots for auctioning purposes. By steering the composition of the lots, a buyer can increase the a...
متن کاملA Heuristic Lotting Method for Electronic Reverse Auctions
An increasing number of commercial companies are using online reverse auctions for their sourcing activities. In reverse auctions, multiple suppliers bid for a contract from a buyer for selling goods and/or services. Usually, the buyer has to procure multiple items, which are typically divided into lots for auctioning purposes. By steering the composition of the lots, a buyer can increase the a...
متن کاملA Comparative Study of Multi-Attribute Continuous Double Auction Mechanisms
Auctions have been as a competitive method of buying and selling valuable or rare items for a long time. Single-sided auctions in which participants negotiate on a single attribute (e.g. price) are very popular. Double auctions and negotiation on multiple attributes create more advantages compared to single-sided and single-attribute auctions. Nonetheless, this adds the complexity of the auctio...
متن کاملCompact bid languages and core-pricing in large multi-object auctions
Combinatorial auctions address the fundamental problem of allocating multiple items in the presence of complex bidder preferences including complements or substitutes. They have found application in public and private sector auctions. Many real-world markets involve the sale of a large number of items, limiting the direct application of combinatorial auctions due to both computational intractab...
متن کامل