Abstract
AbstractThere has been a great deal of recent interest in group technology as a scheme for parts grouping, machine dedication, and shop arrangement, which offers many potential benefits for traditional job shops. Among the benefits cited are a reduction in material handling and set‐up times, less work‐in‐process (WIP) inventory, and shorter flow times. On the other hand, a number of authors claim that many of the advantages believed to be associated with group technology will not occur in practice, due to the inflexibility of machine dedication in group technology.This study investigated a procedure designed to improve performance in group technology shops. The repetitive lots (RL) scheduling procedure capitalizes on the sequence dependency of set‐up times in shops. This procedure scans a queue of waiting jobs, seeking to find a job identical to the job that was just processed on a machine, which should eliminate the need for a machine setup. The truncated repetitive lots (TRL) procedure prevents lots from becoming excessively large by allowing no more than K jobs to be combined. In this study, K was set at five.Computer simulation was used to compare a shop configured as a group technology shop with the same shop configured as a traditional job shop and as a hybrid, which combined features of both group technology and traditional job shops. The group technology shop model had previously been demonstrated to exhibit performance that was inferior to the traditional job shop. Data were gathered on nine variables of interest in alternate simulated years for a period of twenty years, following a seven year start‐up period. The study was designed to investigate two research questions: 1. Will the use of RL scheduling procedures improve performance in group technology shops? 2. Will the use of RL scheduling procedure cause the performance of the group technology shops to be indistinguishable from (or superior to) the performance of traditional job shops?The use of RL scheduling procedures clearly led to an improvement over first‐come, first‐served performance by all shops. This improvement was greatest in the group technology shops (those with dedicated machines). Their machine dedication caused a lower variance of parts types in queues, leading to more opportunities for the combination of lots by the RL scheduling procedures. No substantial differences in performance were found between the RL and TRL procedures; this was probably due to a combination of relatively high K values and relatively low utilization rates.When the three shop environments were compared, using only the RL procedures, major differences among shop environments were found. The shops with dedicated machines exhibited superior performance in terms of set‐up time, machine utilization, and production lot size. However, the traditional job shops showed superior performance in the queue‐related variables: queue length, waiting time, and WIP inventory. These effects combined to lead to an average flow time lower in the traditional job shop than in the shops with dedicated machines. These findings were consistent with the findings from previous studies where repetitive lots procedures were not considered.Although the use of repetitive lots scheduling procedures caused substantial improvement in the performance of the group technology shops, the improvement was not substantial enough to make group technology a viable alternative to the traditional job shop, at least for this shop. Future research in this area should concentrate on the K‐utilization relationship in the TRL procedure, working with other sequence‐dependent set‐up time scheduling procedures and the interaction of repetitive lots procedures with alternate routing.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call