In modern‐day production systems, ever‐rising product variety poses a great challenge for the internal logistics systems used to feed mixed‐model assembly lines with the required parts. As an answer to this challenge many manufacturers especially from automobile industries have identified the supermarket‐concept as a promising part feeding strategy to enable flexible small‐lot deliveries at low cost. In this context, supermarkets are decentralized in‐house logistics areas in the direct vicinity of the final assembly line, which serve as intermediary stores for parts. Small tow trains are loaded with material in a supermarket and deliver parts Just‐in‐Time to the stations lying on their fixed route. This paper discusses the general pros and cons of the supermarket‐concept and treats the decision problem of determining the optimal number and placement of supermarkets on the shop floor. A mathematical model is proposed, an exact dynamic programming algorithm presented, and the validity of the proposed approach for practical purposes as well as the trade‐off resulting from fixed installation and maintenance cost is investigated in a comprehensive computational study.
