AN INTEGRATED MODEL OF STATISTICAL PROCESS CONTROL AND MAINTENANCE BASED ON DELAYED MONITORING IN TWO-STAGE PROCESSES

One of the most important goals of organizations and manufacturing companies is to provide suitable products and services to customers, requiring high-quality processes and keeping them at a desired quality level. In many cases, product quality is low due to equipment deterioration; however, it cannot be figured out until the equipment breaks down. On the other hand, control charts can be used to identify the condition of the process, where the out-of-control state for a quality characteristic means deterioration in the equipment, which is used for the manufacturing purpose. Hence, the statistical process control and maintenance decisions can be combined to form an integrated model that enjoys higher efficiency in reducing costs of quality and maintenance. Furthermore, using delayed monitoring policy to monitor processes is one of the newest research fields in this regard. Delayed monitoring means that processes are in control at the beginning of the process and sampling can be delayed until the pre-specified scheduled time. With a delayed monitoring policy, the total cost of production per unit is expected to be more affordable as the sampling rate decreases; however, it may be lead to an increase in the quality and maintenance costs. Therefore, determining efficient decision variables is important in the model. In this paper, an integrated statistical process control and maintenance model based on delayed monitoring is designed for a two-stage process. By using this procedure, 28 different scenarios are created in which variations in quality and different break-down states are considered. barX residual mean control charts have been used for monitoring purposes. In order to integrate statistical process control and maintenance, a model is proposed such that the expected cost per time unit of manufacturing is minimized by using a genetic algorithm. To evaluate the performance of the proposed method, an illustrative example is presented. In addition, sensitivity analysis of some parameters of the proposed model is carried out. The results show the appropriate performance of the proposed model.