An Optimal Coalition Formation among Buyer Agents Based on a Genetic Algorithm

Group buying is a form of electronic commerce that is growing quickly. There are many group buying sites on the Internet. Group buying is a commercial transaction in which the unit price of goods changes with the number of buyers, and a buyer can purchase goods at a low price if many buyers participate in group buying. There are several group-buying sites that are selling similar (or the same) goods. Buyers are often distributed among these group-buying sites nonoptimally. If we can optimally allocate buyers to several group buying sites, all buyers can buy a good at a lower price. The aim of this paper is to solve this optimal allocation problem by a Genetic Algorithm. Our method can effectively avoid the growth of fatal genes. In experiments, we compared our method with an exhaustive search algorithm and a brute search algorithm. The experimental results show that our algorithm can optimally allocate buyers in an efficient time.