Search

In this paper, we conduct a case study of a company that collects and processes used home appliances and sells them as new recycled resources based on the Home Appliance Recycling Law. The target company earns sales by selling multiple resources to multiple sales destinations with different selling prices. There are various restrictions on such sales, which creates complexity. Due to this complexity, the sales amount cannot be maximized using only the intuition and experience of employees. Therefore, we propose a model that maximizes the sales amount obtained from selling each resource. As a result, using the proposed model, different sales amounts and sales destinations of each resource were obtained, achieving better results. Specifically, compared to the current total sales amount, 3.6% better results are expected. Furthermore, as a sensitivity analysis, experiments were conducted in which the credit constraints of each sales destination were varied. It was found that risk management can be performed without affecting the objective function value by setting credit appropriately. [ABSTRACT FROM AUTHOR]

In recent years, various networks have come to exist in our surroundings. This not only includes examples such as the internet and airline routes that can be thought of as networks; protein interactions are also networks. A network is defined as an arrangement of nodes (points) and links (lines). An "economic network design problem" can be discussed by assuming that a vertex is an economic player and that a link represents some connection between economic players. An economic player can mean an individual, an association, a city, a nation and so forth. A connection between economic players signifies items such as a transport link, a telecommunication net or an economic regional alliance. In this paper, an "economic network design problem" is modeled based on Jackson and Wolinsky's connections model, and the utility function of the network is defined. In the field of economics, generally, the game theory approach is adopted for analysis of network formation. However, we encounter difficulties when a network has many heterogeneous nodes. Therefore, we suggest that the optimization of an "economic network design problem" can be solved using evolutionary computation. As an example of an "economic network design problem," we consider an optimization problem of a network between regions in West Japan. In addition, the priority of the network maintenance is clarified via the optimum network obtained in the simulation. [ABSTRACT FROM AUTHOR]