Industry and academia are both making efforts to realize a sustainable society; an important part of these efforts is to ensure the sustainability of the supply chains that support our daily life. Sustainable supply chains are more complex than traditional supply chains, and they involve a combination of multiple issues. Multiple plans must be used to deal with these issues. However, these plans often conflict with each other. To manage a sustainable supply chain, an integrated approach is needed to operate multiple plans for multiple issues.This paper introduces a research concept for integrated sustainable supply chain management using a multi-agent system. An agent executing a plan for an issue autonomously negotiates with other agents and avoids conflicts. Linear physical programming used for negotiation balances agents’ utility and ensures that all plans function well. Through this research, we provide an example of the simultaneous operation of multiple plans in a sustainable supply chain, aiming at the social implementation of sustainable supply chains.

As environmental problems become more apparent, manufacturers need to balance environmental considerations with economic activities. This is where closed-loop supply chains are gaining attention. However, in addition to demand fluctuations, which are a problem in conventional supply chains, a circular supply chain is unstable in terms of supply, where end-of-life products are collected and reused. This destabilizes not only excess inventory and shortages but also production resources, such as manpower, facilities, and raw materials. This study focuses on the stabilization of the manufacturing system in a closed-loop supply chain. To confirm the dynamic changes in the manufacturing system, we designed a simulation model of a closed-loop manufacturing system and conducted numerical experiments under several scenarios, taking the variation of manufacturing quantity per unit period as an evaluation measure of stability. After showing that unplanned remanufacturing destabilizes the recovery of reusable end-of-life products, we demonstrate that the manufacturing system can be stabilized by appropriately limiting the amount of remanufacturing. However, excessive limits reduce opportunities for remanufacturing end-of-life products and generate adverse economic and environmental impacts. To determine appropriate restrictions, it is necessary to consider the product currently in use by the customer as a virtual inventory and to consider factors such as the quality of the products in the virtual inventory. In the future, we plan to study a system that can dynamically manage remanufacturing quantities based on the status of virtual inventories.

Environmental issues and increasing global demand have led to calls for sustainable production and consumption. Products deemed insufficiently environmentally friendly are beginning to be removed from the market, and manufacturers must achieve social responsibility and economic benefits. However, in a conventional linear supply chain, an increase in sales volume directly leads to an increase in waste and resource consumption, resulting in a trade-off between the environment and the economy. Closed-loop supply chains resolve trade-offs, achieve corporate social responsibility, and enhance competitiveness. Closed-loop supply chains must recover used products with a high degree of uncertainty. Uncertainty in the quantity, quality, and timing of used products can cause serious problems such as the bullwhip effect in the dynamic behavior of the supply chain. This study aims to analyze the dynamic impact of used product uncertainty on a closed-loop supply chain and propose ways to deal with it. Products deteriorate with their use and eventually become worthless for recovery and remanufacturing. This study confirms that, even when there is no demand variability and lead time, the uncertainty of used products, especially qualitative uncertainty, causes large variability in the closed-loop supply chain. The results of the numerical example using the model proposed in this study show that the policy of limiting the quantity of remanufacturing can mitigate the variability. Furthermore, the appropriate limits vary depending on the characteristics of the product, such as its lifetime. This suggests the need to control the amount of remanufacturing for the sustainability of the closed-loop supply chain rather than remanufacturing as much as possible, to reduce environmental impact.