With the emphasis on revitalization of small and medium-sized enterprises, the importance of supporting systems for consulting, which supports SMEs to solve problems that SMEs are difficult to deal with, is increasing. Since problem solving of client companies and service proposals depend on the experience and intuition of consultants, support methods are required to provide stable quality of service. As a first step in the support method, this study focuses on text data that describes communication with client companies and predicts the occurrence of problems in client companies using text mining.

Among the service industries, the food service industry has low labor productivity. In this research, in order to improve the productivity and to achieve optimality of resource allocation and improvement of added value at the same time, the shift planning method based on combinatorial auction is proposed. In the shift planning method, it is necessary to simultaneously decide "when each employee work" and "what task each employee is in charge of", while securing necessary personnel to provide more than a certain level of service and aiming at not to work as much as possible.

In recent years, trade balance is very important for current account balance for each country, and a profit gained by the trading plays an essential role for surplus of the current balance. In our group, the trade model in the cyber field is constructed using a malti agent systems and the effects due to economic of several countries are evaluated. The purpose of this paper is to verify the change of the trade profits and the export volumes in the case of tariff elimination. The computer experiments are performed and the changes are discussed.

　World is growing very fast and technological advancement creates system after system, increasing interdependence and overall complexness. In order to handle this changes new systems approaches are necessary.
Objective of research: Development of multiscale simulation modeling approach and its application to electricity market analysis Hypothetical model consists of triple scale.

• Macro scale: high abstraction, monitors macroeconomic variables
• Meso scale: identification of optimal energy mix, sets up electricity price
• Micro scale: hour to hour micro level decisions about demand determination

This research focuses on the negotiation protocol between multiple Manufacture Agents (MA) and multiple Material Supplier Agents (MSA). A hierarchical-game based negotiation protocol is proposed. It is a three-layer game, where the first layer game is based on the results of the second and the third layer games. Two-person game is used to find the optimal trade partnerships to maximize the whole profit of Supply Chain Network (SCN) in the first layer. The second layer games aim to find all the possible coalitions using cooperative game. However, they are not necessary for all the time. They exist only at the situation where the order of MA exceeds the ability of MSA. Then, the third layer games are used to determine the final strategies between MAs and MSAs or all the possible coalitions found in the second layer games. Stackelberg equilibrium is introduced to resolve the conflict of the interests of the two sides. What we should pay attention to is that the second and the third layer games are nested inside the first layer game. Simulations are provided to verify the effectiveness and the feasibility of the proposed protocol. The optimal setting of the proposed protocol under given condition is obtained by the performance analysis.

Central Kitchen in the food service industry is a labor-intensive workplace where preparation by hand generates additional values. Thus, the production efficiency greatly depends not only on the flow of goods but also on the flow of people. In this respect, this research aims to find the optimal facility layout based on the method which combines the Genetic Algorithm and computer simulation.

Recently renewable energy has attracted attention, although its power generation capacity is unstable under dynamic environment. In this respect, the smart grid technology is expected to help control power distribution. This research proposes a negotiation mechanism to derive an optimal power distribution planning in a region using the Market-Oriented Programming (MOP) which contains a multi-agent social negotiation based optimization method. Effective dynamic pricing strategy is expected by this research.

This research aims to build a Particle Swarm Optimization (PSO) method which autonomously changes the settings for each solving problem. The structure of the solution space of an optimization problem is presumed based on the information acquired during search of the problem. According to the presumed structure, PSO parameter settings will autonomously change.

This research focuses on shop layout design of underground shopping streets which is considered to affect purchase behaviors. Agent simulation is applied to the shop layout design for the developers in the aim of attracting more purchasers to underground shopping streets.