Proceedings of the 2010 International Conference on Industrial Engineering and Operations Management Dhaka, Bangladesh, January 9 – 10, 2010 Impacts of Common Components in Multistage Production System under Uncertain Conditions M. A. Wazed, Shamsuddin Ahmed, and Nukman Yusoff Department of Engineering Design and Manufacture University of Malaya (UM), 50603 Kuala Lumpur, Malaysia Abstract The work desires: i) to determine the optimum level of batch size in bottleneck facility and ii) to analyze the effect of common components on work-in-process (WIP) level and cycle time in a multistage production system under uncertainties. The uncertainty is created by machine breakdown and quality variation. Few simulation models are developed based on a live case from a company. The models are verified and validated with the historical data from the company and by face validity. Taguchi approach for orthogonal array is used in designing experiments and these are executed in WITNESS. It is observed that the variation in level of common component in the system has significant impact on the production WIP level and cycle time. The main contribution of this research is determination of the optimal level of batch size in a bottleneck resource under the uncertainties. This approach can be generalized to any multistage production system, regardless of the precedence relationships among the various production stages in the system. Keywords Simulation, Quality, Machine breakdown, Work-in-process, Cycle time 1. Introduction The classical lot sizing model assumes the output of the production process is of perfect quality. However, in real manufacturing system, nonconforming items may produce as time goes. These nonconforming items need to be screened out. The presence of defective product motivate in a smaller lot size. Optimum lot size for each stages even more complicated in multistage production system when cycle time for each stage is different. The number of defectives may vary in multistage production system where the products move from one stage to another. Depending on proportion of defective items, the optimal batch sizes in the stages also varies. However, small batch size may reduce the productivity and stock out and this increase the total expected cost. Thus, an optimum lot size must be obtained when quality is stochastic. Multi-stage production planning is a system which transforms or transfer inventories through a set of connected stages to produce the finished goods. The stages represent the delivery or transformation of raw materials, transfer of work-in-process between production facilities, assembly of component parts, or the distribution of finished goods. The fundamental challenge of multi-stage production is the propagation and accumulation of uncertainties that influences the conformity of the outputs [1]. The present study is concern with such a multistage system and simulation is chosen to analysis the objectives. A simulation model is a surrogate for experimenting with a real manufacturing system. It is often infeasible or not cost-effective to do an experiment in a real process. Thus, it is important for an analyst to determine whether the simulation model is an accurate representation of the system being studied. Further the model has to be credible; otherwise, the results may never be used in the decision-making process, even if the model is “valid” [2]. Few simulation models are used to analyze various effects of uncertain factors namely machine breakdown and quality variability. Machine breakdown means the failure or stoppage of machine(s) for unknown reason(s) and a representation of interruption in the process [3]. It wields a reduction of capacity level and delay the release of products or