This paper analyses the tool switching problem in the dynamic environment, where parts to be processed on a flexible machine arrive randomly. As the total number of tools required to Process a set of parts on the machine is generally larger than the available magazine storage capacity, tool switches between parts are usually necessary. We assume that tool switching must be made just before the processing of the parts. Since the time required for tool switches can be significant relative to processing time and cause the processing of parts to be delayed, it is desirable to minimize the number of tool switches. Therefore, we present one heuristic algorithm to minimize the total number of tool switches and the t)reposed heuristic is compared with the KTNS (Keep Tool Needed Soonest) policy on randomly generated problems.

Abstract
 I. Introduction
 II. Problem description
 III. A heuristic for the dynamic tool switching problem
 IV. Computational results
 V. Conclusions
 References

• Song Chang-Yong
• Shinn Seong-Whan