نشریه مهندسی و مدیریت کیفیت
سال یازدهم شماره 3 (پاییز 1400)

One of the most widely used tools in the rapid detection of assignable causes is control charts. Given the importance of economic costs, Duncan proposed the first economic model in the presence of multiple assignable causes in order to reduce the economic costs of the quality cycle. In his model and all the economic designs derived from that, assumed that after the occurrence of an assignable  cause, process is free from the occurrence of other assignable causes. In this paper, after criticizing previous models for incorrect and unrealistic use of this assumption to calculate the average cost per unit of quality cycle time, a realistic economic-statistical design in the presence of multiple assignable causes for economic-statistical design of  X-bar control chart is presented. The numerical results of our model show well that in the previous models; the average cost per unit time of the quality cycle is severely underestimated compared to the actual value and with increasing the Weibull distribution shape parameter, the probability of this assumption is greatly reduced. Therefore, it is suggested that in order to eliminate the shortcomings of the economic design of various types of control charts with multiple assignable causes in future research, they should be redesigned based on our model.

This paper for the first time in the statistical design of adaptive control charts, deals with the statistical design of univariate control chart X ̅ under the Weibull shock model. practically, the use of shock models with more flexible risk rate functions in the statistical design of adaptive control charts is closer to reality. This study shows that diagram X ̅-VRS under Weibull shock model with non-uniform sampling intervals and variable sample size, compared to diagram X ̅ under Weibull shock model with non-uniform sampling intervals and fixed sample size (FRS), Detecting average changes is faster and performs better.In this model, with increasing changes in the mean, the rate of detection of changes in the mean increases and the value of h_1 increases and the ANF decreases. Also relatively large changes (𝛅≥2), lead to a relatively small sample size (n_2≤ 10) and smaller changes (2 ≤ 𝛅 <0.25) lead to a larger optimal sample size (〖"11≤n" 〗_2 "≤14" ).

Expanding product variety helps customers to find products that perfectly fit their individual needs. Therefore, companies are looking for ways to better manage and variety procedures in the product and a design that is compatible with variety can be considered a competitive advantage for the organization. The present study aims to evaluate engineering designs with a variety approach, has presented an optimization model that examines engineering designs in terms of two parameters, the amount of changes required by the design to standardize now and future manufacturers' efforts to redesign the components. The parameter of the amount of changes required by the design for standardization now is obtained with the help of the generational variety index and the coupling index and the parameter of future manufacturers' efforts to redesign the components with the help of the commonality index. Applying the model developed in the present study, as well as determining the allowable components in the standardization priority and the amount of effort in the future will lead to the selection of engineering design that can reduce the cost of product development, redesign efforts and market time. The model is developed in the present study has been implemented on one of the fuzzy array radars of Iran's electronics industry and has determined the optimal engineering design of the desired radar.

The production of low quality drugs in addition to wasting resources, leads to risks to public health. Quality risk management can ensure the high quality of the drug for the patient by controlling quality risks during the product manufacturing process. The present study aimed to identify and analyze the particle contamination risk of the vial product at Daana Pharmaceutical Company by implementing quality risk management process based on the ICH Q9 guideline. For this purpose, in order to identify the risk, the failure modes and effects analysis (FMEA) method and to identify the root causes of failure, the "Why" root cause analysis (Why RCA) technique and to analyze the risk, a combination of the failure modes, effects and criticality analysis (FMECA) and Bayesian Belief Network (BBN) analysis methods were used. Comparison of the results of the FMECA and BBN-FMECA approaches showed that at Daana Company, risk analysis with both methods gives similar results. Also quality of raw material powder and primary packaging materials, personnel performance and air conditioning monitoring system are identified as the most important causes of product contamination risk

This article deals with modelling and analysis of the competing risks for a one-shot device under a constant stress accelerated life test. In a reliability analysis of a device, it is important to be able to identify the main causes of failure. Therefore, a competing risk model is generally used. We consider this model in two modes: observed and masked causes of failure. The data obtained from one-shot device testing are missing in fact. For this reason, the EM algorithm along with the Fisher scoring method are used to estimate the model parameters. An accelerated life test is also used to shorten the time and cost. In addition, in order to accurately estimate the product reliability, the test design is finally optimized. Based on the simulated study, it is concluded that the EM algorithm and the bootstrap confidence interval are more accurate than the other methods. Also, shortening the test length leads to achieve an optimal test design.

Todays, the performance of a process or the quality of a product in conditions of uncertainty and under distributions from the exponential distribution family is evaluated by a fuzzy communication model with binary data called fuzzy generalized linear profiles. Generalized linear profiles are a type of nonlinear profile in which process observations follow the Bernoulli or binomial distribution. In this research, approaches In order to monitor fuzzy generalized linear profiles in phase 2, we propose. The main purpose of this paper is to monitor the fuzzy statistical process to detect the time of occurrence of changes in processes as fuzzy change point and based on the principle of maximum likelihood (MLE). Is based on fuzzy observations. Performance of the proposed method for monitoring fuzzy generalized linear profiles based on the probability of an out-of-control signal using the fuzzy control diagram (FEWMA) and then estimating the fuzzy change point for the simulated data and real data.