A mathematical model in determining an optimal policy for an N-stage production line with random rework and yield loss

• Main Authors: Say, Arlene, Tiu, Pamela, Ty, Jeanne
• Format: text
• Language: English
• Published: Animo Repository 1997
• Subjects: Mathematical models; Stochastic processes; Scheduling (Management); Production engineering; Random operators; Systems engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/8311
• Institution: De La Salle University

Scheduling of jobs in a facility in a production line is important particularly if the workstations perform different types of operations requiring machine changeover. Determining an operational scheduling policy would therefore increase line productivity and minimize unnecessary production costs. This study focuses on critical workstations undergoing in-process 100% inspection, wherein the output would either be good items, yield loss or rework items. The good units resulting from such a facility forms transfer batches before they are moved to the next facility. An output of a transfer batch in the final stage of the line is, in the same way, required by the system and these would also affect the transfer batches at the end of the each work facility. The rework items, on the other hand, are placed in a buffer and rerouted to the same workstation for a one-pass rework operation. Items reworked are then classified as either yield rates or yield losses. Holding cost and switchover cost are high, thus, finding the critical value for processing reworks would serve to balance these costs. Too small a critical value would result in frequent switchover which is a considerably high cost. Too big a critical value would then result in increased holding cost for the rework jobs. Hence, there is a need to come up with an operating policy of when to process these rework jobs. The model was formulated on the basis of one cycle (regular operation to rework operation to regular operation)