Abstract
۱٫ Introduction
۲٫ Problem statement
۳٫ Stochastic dynamic programming
۴٫ Model formulation
۵٫ Illustrative example
۶٫ Conclusion
Acknowledgements
References
Abstract
Process quality planning should establish the quality control plan to achieve the desired quality level with the minimum quality cost (appraisal and failure costs) for the final product. This plan sets out the critical quality variables, the control stations in the process, and the control method at each control station. The quality costs associated with quality control and defective products can be greater than or less than ideal regarding the required quality level. The purpose of this paper is to provide a stochastic dynamic programming model for designing the quality control plan in a manufacturing process, which allows obtaining the desired level of control with the lowest cost. Inputs to the model are, in particular, control stations in the process, levels of quality, control methodologies (no control, statistical process control, 100% inspection), probabilities of changing the quality level and quality costs. The output of this model is the quality control plan that satisfies the desired level of quality at the lowest cost. This plan establishes the control stations, the methodology used in each control station, the desired quality level for the final product, and the estimated quality costs. Finally, an illustrative example based on a manufacturing process demonstrates the applicability of this approach and several considerations are reported about future research directions.
Introduction
The combination of high quality products at low cost has been a critical issue for manufacturers to remain currently competitive in global markets. Companies use a variety of methodologies (sampling inspection, 100% inspection, reinspection, control charts …) that allows them to achieve the desired quality levels at competitive costs. Such methodologies focus on operations where there is a greater probability of occurrence of failures or its impact is more significant. In this way, an integrated approach to the quality of the manufacturing system is not done, resulting in a misuse of human, financial and material resources and increased quality costs. This situation is more relevant with the increasing rate of introduction of new products in companies [1]. Manufacturing systems are generally composed of several workstations (WS) or stages, in which raw materials pass through various operations and are transformed into finished products. This type of systems is called multi-station (or multi-stage) manufacturing systems (MMS) [2]. In MMS, each WS will produce a proportion of defective items [3]. To economically maintain the product quality level is a critical issue for the MMS, in which each station may inevitably shift to the out-of-control condition resulting in higher nonconforming rate and larger quality loss [4, 5].