제안 방법

• In an attempt to improve the performance of the supply delay, we proposed the Minimum Distance (MD) rule that aims to control part inventories toward to a proper level. A comparative analysis on the performance of the traditional rules and the MD rule is provided via a simulation model. The new rule shows a significant improvement over the traditional rules with respect to the mean supply delay.
• This rule demonstrates that the supply delay can be maintained at a minimum level only when the inventory variance is ultimately minimized over the scheduling horizon. Finally, we provide a comparative analysis on the performance of traditional rules and the new rule proposed in this paper via a simulation model.
• For each experiment set, the performance measures of flow time, supply delay, and part lead time were investigated. The numerical results are presented in terms of mean and standard deviation values.
• In this paper, we are concerned with product scheduling by which the inventory control of disassembled parts can be incorporated into a priority rule to improve the part lead time performance. However, the inventory control cannot be done in an independent manner because of the distinctive disassembly sequences.
• In their next study, the algorithm is extended to two companion algorithms, the core algorithm and allocation algorithm (Taleb and Gupta, 1997). The core algorithm determines the amount of used products required to be disassembled over the planning horizon, and the allocation algorithm determines when and what products should be disassembled for each period, considering the disassembly time and order interval. Unfortunately, Taleb and Gupta ignored the characteristics of the disassembly environment, including dynamical product arrival, disassembly operation, and order interval.
• The important factors of this simulation study are service levels of parts, job structures, and priority rules. Three levels of the part service are simulated in the experiment: 85% and 95% for all part types.
• A total of eight different types of parts that are referred to with lowercase letters a, b,…, g and h are released through the predetermined release sequence for each product type. Therefore, the disassembly shop simulated in this study is designed to have eight different types of disassembly machine to disassemble each part. Meanwhile, Holthaus and Rajendran (1997) demonstrated their job-shop simulation with only six machines, claiming it was adequate to represent the complex structure of a job shop.
• No priority rules for product disassembly scheduling have been reported in the existing literature. To show the effectiveness of the proposed MD rule, we examine the scheduling strategies of first-in, first-out (FIFO) and shortest processing time (SPT). The processing time of a product is measured based on the time required for separating all parts from the product.