Machine buffer sizing based on the Theory of Constraint principles and Tabu Search techniques

In the last few years firms efforts have been directed toward the improvement of manufacturing cycle, lead times reduction and Quality Assurance, due to the ever increasing time-to-market and quality requirements. In this context to achieve better results in terms of shop floor scheduling and lead time reduction, many scheduling approaches have been developed, but most of these fall short of achieving good results (Buzzacott 1968, Davis and Taylor 1978, Conway et al. 1988 and others) because they do not identify clearly where and how much inventory should be employed (for example Davis and Taylor propose a heuristic for determining work in process inventories in a flow line). This paper deals with problems of WIP inventory management, describing an approach based on the combinatorial problem related to WIP size and location on shop floor, which is a function of buffer sizes in each operation of the production process. The proposed model has been tested in an aeronautical firm where we have achieved a relevant growth of the output rate level using four work in process buffer with a total of 50 batches inventory.