Due to growing concern about environmental problems, it is becoming important for manufacturers to add more value while causing less environmental impact. In order to reduce the environmental impact of products over their entire life cycle, adequate reuse and recycling of products and their components are quite promising [1, 2]. In this context, it is quite important for manufacturing firms to establish efficient closed-loop manufacturing systems (CMS)  in which products are made from used components and materials as well as new ones. Some firms have successfully established quite efficient CMS from both environmental and economical viewpoints. CMSs for one-time-use cameras , photocopying machines , and automobile components  are typical examples.
However, establishment of an environmentally and economically efficient CMS is not easy, mainly due to high uncertainty associated with the return flow of post-use products. Since product usage conditions and lifetimes differ from user to user and cannot, in general, be controlled by manufacturers, there are significant fluctuations in the quality and quantity of product return flows [7, 8]. In addition, the return flow of post-use products may contain different product models in different conditions, each of which requires different remanufacturing operations (e.g., some may need cleaning and inspection while others may need disassembly into their components). Therefore, CMS should have higher flexibility and redundancy than conventional production systems to adapt these significant fluctuations.
Both of these requirements are quite expensive to meet. Flexible machines and labours are generally more expensive (sometimes less effective) than fixed purpose ones. In addition, the differences in necessary operations for each used product need frequent reprogramming and set up for manufacturing equipment. This hinders the automation of CMSs and results in higher operation cost, especially in developed countries where labour cost is expensive. The high redundancy in production resources also leads to their less efficient utilization and causes higher investment cost than conventional ones.
In order to solve these problems, many studies have been conducted in recent years. Examples include, Holonic Manufacturing Systems (HMS) , Biological Manufacturing Systems (BMS) , cellular manufacturing systems , and SOCRADES (Service Oriented Cross-layer infRAstructures for Distributed smart Embedded deviceS)  based on Service Oriented Architectures (SOA) . Some of these [9–11] focus on the development of completely new conceptual (sometimes ideal) flexible manufacturing systems, while others [12–14] concentrate on enabling technologies (e.g., XML-based communication protocols for embedded devices and semantic webs for realizing SOA).
However, most of the studies assumed complete replacement of existing systems, which might require prohibitive investment at the beginning. There is a lack of systematic and practical methods for improving the flexibility of existing systems by gradually introducing these concepts. This is a major reason for that many of these concepts have not spread widely into industry.
The objective of this study is to propose a strategic decision making method for designing environmentally and economically efficient CMS while maintaining the flexibility and the redundancy to adapt the significant fluctuations in product return flows. Especially, this paper deals with the investment reduction of a CMS through effective sharing of its resources across multiple production systems.
To this end, we introduce a transferability benefit index (TBI), the ratio of the benefits to difficulties, to identify the most promising resources for sharing among multiple production systems. We also provide a simplified example calculation to illustrate the method and discuss its result and the future development needs of the methods.