The proliferation of the Internet and communication technologies and applications, besides the conventional retailers, has led to a new form of distribution channel, namely home sopping through the telephone, TV, catalog or the Internet. The conventional and new distribution channels have different transaction costs perceived by the consumers in the following perspectives: the accessibility to the product information, the traffic cost and the opportunity cost for the time to visit the store, the possibility of ‘touch and feel’ to test the quality of the product, the delivery time and the concern for the security for the personal information. Difference in the transaction costs between the distribution channels results in the different selling prices even for the same product. Moreover, distribution channels with different selling prices necessarily result in different business surpluses. In this paper, we study the multilateral bargaining strategy of a manufacturer who sells a product through multiple distribution channels with different transaction costs. We first derive the Nash equilibrium solutions for both simultaneous and sequential bargaining games. The numerical analyses for the Nash equilibrium solutions show that the optimal bargaining strategy of the manufacturer heavily depends not only on the degree of competition between the distribution channels but on the difference of the business surpluses of the distribution channels. First, it is shown that there can be four types of locally optimal bargaining strategies if we assume the market powers of the manufacturer over the distribution channels can be different. It is also shown that, among the four local optimal bargaining strategies, simultaneous bargaining with the distribution channels is the most preferred bargaining strategy for the manufacturer.

Recently, transaction data is accumulated everywhere such as bank and IT company. Association analysis methods are usually applied to analyze transaction data, but the methods have several problems. For example, these methods only consider one-way relations among items and cannot reflect domain knowledge to analysis process. In order to overcome defect of association analysis methods, we suggest a transaction data analysis method based on probabilistic graphical model (PGM) in this study. The method we suggest has several advantages as compared with association analysis methods. For example, this method has a high flexibility, so it can give a solution to various probability problems regarding the transaction data and can consider various relationships among items.