An investigation of the impact of a flexible manufacturing island on the operating performance of a manufacturing system

A number of investigations point to problems associated with the introduction of various types of factory automation, particularly FMS. Research indicates that the success of the implementation in such automation is often hindered by lack of integration of the new technology with the existing conventional system. This study investigates operational issues associated with the introduction of a flexible manufacturing work center (Flexible Manufacturing Island or FMI in this research) into an otherwise conventional system. The primary objective of the research was to investigate the impact on system performance associated with the introduction of a flexible manufacturing island. Also, the study compares the relative impacts on system inventory holding cost caused by installation of a single FMI at different locations within the system. Finally, the study investigates the sensitivity of these impacts on inventory holding cost to changes in (1) system utilization, (2) the ratios of setup times to run times in the conventional work centers, and (3) the rates of increase in holding costs for items as they move through the system.;The research methodology is explained by using an example. Some of the conclusions of the research for the system studied are: (1) Conversion of a work center into an FMI in an otherwise conventional system reduces inventory holding cost for the system. (2) Conversion of a bottleneck work center into an FMI has the greatest impact in reducing inventory holding cost. (3) Conversion of a work center at the back-end of the system into an FMI has greatest impact in reducing inventory holding cost. (4) Inventory holding cost is relatively sensitive to capacity utilization rate, and the relationship is non-linear. (5) Inventory holding cost is relatively insensitive to the rate of increase in holding cost per unit. (6) The relationship between inventory holding cost and the ratio of setup time to run time is approximately linear.