Heuristic methods for Q-time bottleneck dispatching

This paper presents the design of an efficient heuristic method for q-time bottleneck dispatching (QBD) to achieve the bottleneck utilization and minimum of work-in-process (WIP) level in order to reduce the hold lots in q-time start step and prevent q-time violations from quality issues. The core concept of this method consists of linear programming problems of operation constraints such as bottleneck machine's capacity planning, queuing WIP limitation, flow-in and re-entrant WIP pre-allocation. Base on the theory of constraints (TOC) and heuristic method, the mathematical model was implemented into real time dispatching system (RTD) to simulate and systematize machine loading and control q-time variability. This method improve the q-time hold lots by 75% and the average q-time violations cases was shortened by 31% from bottleneck extra queuing WIP reduction.