A simulation study to compare Minimum Inventory Variability Policies (MIVP®) and First-In-First-Out (FIFO) algorithm

Abstract

Minimum Inventory Variability Policies (MIVP®) for product release and resource scheduling developed by ACADZ, Inc. is one of many existing dispatching rules used in manufacturing and research today. The purpose of these two research studies was to confirm that the MIVP® K-Step Ahead® policy is more efficient at resource scheduling than First-In-First-Out (FIFO) in terms of average cycle time and cycle time variance in a stable semiconductor factory. FIFO is used as the baseline for comparison of a two-product DRAM semiconductor factory dataset provided by SEMATECH. The AutoSched/Automod® simulation software tool was chosen to model the system. The research results suggest significant improvement in average cycle time, and cycle time variance when the factory was loaded to within 95% capacity but not in cycle time variance when the factory was loaded at 85% capacity when using MIVP® K-Step Ahead®.