Journal of Quality Engineering and Management

Abstract Dispersion is the enemy of quality and an inherent and integral part of manufactured products. Therefore, identifying critical components and their contribution to total variation is an important engineering task, and measuring it in general without the most restrictive assumptions is very complex. The present article presents a systematic approach that can identify the contribution of each component to the total variability in a complex system, in proportion to the type of mechanism of action of the components. The index introduced in this study determines the contribution of the components and can be used as a measure of the criticality of the components of a system. The application of the proposed method in this article does not require any assumptions regarding the linearity of the functional function of the components or the normality of the statistical distribution of the quality characteristics, and includes the analysis of all systems with uncorrelated components. The proposed solution can be used as a powerful tool in the analysis phase of Six Sigma and Lean Six Sigma, and with its help, it is possible to prioritize and policy the use of resources in order to reduce process dispersion. To understand the proposed method in more detail, two well-known examples in industrial engineering are described.

Abstract In the statistical control of multivariate processes, two or more quality characteristics must be controlled simultaneously. In controlling such processes, two main goals must be achieved. The first goal is to detect out-of-control conditions and the second goal is to identify the quality characteristics that cause the deviation when an out-of-control condition occurs. In this research, ways to achieve the first goal are investigated and methods for monitoring the multivariate variance-covariance matrix in phase 2 are presented. The main goal of phase 2 is to quickly detect shifts. In this paper, two methods for monitoring the multivariate variance-covariance matrix in phase 2 are presented and the shift in one of the quality characteristics of case 1 (average trail length) and the detection of ARL are investigated. Simulation results show that the proposed methods reduce the out-of-control condition more quickly.

Abstract The sign chart is one of the most common nonparametric charts used to control the centrality of processes with unknown or non-normal distributions. Considering the effect of measurement error on the performance of control charts, this paper examines the effect of measurement error on the performance of the sign chart using an additive model. For this purpose, a simulation program has been developed that calculates the average length of the sign chart sequence for three different distributions and in two states of awareness or ignorance of the existence of measurement error. The simulation results show that the performance of the sign chart for all three distributions and in both states is weakened by the effect of measurement error, and with an increase in the variance of the measurement error, the effect of the error on the performance of the chart increases. Also, the effect of increasing the number of measurements on reducing the effect of measurement error in the sign chart has been investigated. The results show that although using this method when aware of the existence of measurement error has a positive effect on the performance of the chart, if unaware of the existence of the error, this method weakens the performance of the sign chart.

Abstract Traditional control charts are based on the basic assumption that process data are sequentially independent of each other and have a normal distribution. However, in many real-world cases, including chemical and continuous processes, this basic assumption does not exist and there is a kind of autocorrelation between the data collected from the process. The use of traditional control charts in autocorrelated processes is unreliable and increases false alarms. One of the methods developed to control autocorrelated processes is to identify the structure of the process time series and use the residual values ​​to control the process. In this paper, a model based on neuro-fuzzy adaptive systems is designed to identify the structure of the time series and use the prediction. Finally, it is AR(2). Also, residual control charts based on this system for second-order autoregressive data using simulated data, the efficiency of the proposed method in the weighted moving average chart and for different degrees of correlation are evaluated, and it is shown that the proposed method has very good efficiency for data with high correlation.

Abstract The current trend in various industries acknowledges that establishing a system with the ability to quickly refer to the level of product failures or estimate its reliability is a necessity for every industry. Reliability is doubly important in military industries. One of the products of the military industries is anti-aircraft shells that are used against enemy threats, and the failure or failure to act on time of this product can cause irreparable damage, which increases the importance of this product. In this case, it is in the design stage and there is no previous or experimental data available, the lack of data is considered the main problem, and to solve this problem, we use Bayesian networks, and due to the lack of knowledge of the reliability of components in the design stage and the lack of sufficient and accurate knowledge of experts about the reliability of components, a basis is considered for the reliability of each member, and fuzzy theory is used to obtain reliability. To obtain reliability using fuzzy Bayesian networks, we first draw the product fault tree and by converting the fault tree into Bayesian networks, the product reliability is estimated.

Abstract Successful organizations continuously evaluate the quality of the company's products and services with other competitors, but self-assessment is more important from a quality perspective. This research, by using the "performance measurement based on network auxiliary variables" model, helps managers to self-assess the quality of products. The proposed model in this research is examined in the form of a case study (Moghan Wire and Cable Company) and the quality performance of the company's products over time is analyzed. The results of this research indicate the effectiveness of the proposed model.

Abstract Business excellence initiatives help organizations develop and increase management capabilities in order to achieve high performance and greater competitiveness, and management commitment to implement excellence models and self-assessment based on them is an important initial stage of the organizational excellence process. The present applied research seeks to identify and classify all its components in the implementation of business excellence models by examining the concept of management commitment. In this regard, by reviewing the research literature and conducting semi-structured interviews with experts, 92 components that form the concept of management commitment were identified and counted, and validated using a questionnaire and their importance was assessed. Based on the implementation of exploratory factor analysis, the components of management commitment were categorized into 4 main factors: ownership of organizational excellence, improvement programs, membership in self-assessment teams and creating integration, and creating and developing a culture of excellence. By identifying and categorizing the components of management commitment, this research lays the groundwork for creating a broader understanding and awareness for managers of their roles and responsibilities in implementing excellence models and increasing the effectiveness of the organizational excellence process.