Journal of Quality Engineering and Management

Abstract Nowadays, many products are the outputs of multi-stage processes. In such processes, the stages are often interdependent, meaning that the quality of the product in a particular stage depends not only on the quality in the current stage but also on the quality of the product in the previous stages. This phenomenon is referred to as the cascading property of multi-stage processes. The existence of this property and the lack of attention to it can lead to errors in interpreting the control charts used in the stages. Therefore, in the literature on multi-stage process monitoring, methods have been proposed to reduce or eliminate this problem. On the other hand, in some cases, the quality of a product is described by the relationship between a response variable and one or more independent variables, known as a profile, which can be the result and output of a multi-stage process. Since fewer studies have been conducted on monitoring profiles resulting from multi-stage processes, this paper investigates the effect of the cascading property on the monitoring of simple linear profiles in a two-stage process, based on the average run length criterion using simulation, and also examines the effect of inter-stage dependency on the estimation of the profile parameters of the second stage.

Abstract Since the beginning of human existence, humans have sought ways to reduce risks and threats in their living and working environments and to make them safer. Accidents such as airplane crashes, factory explosions, and similar events, which have resulted in extensive human and financial losses, have further emphasized the importance of accuracy and focus on methods for reducing the occurrence of such incidents. Moreover, with the rapid advancement of technology and the increasing complexity of systems, the severity of damages and consequences arising from accidents has increased. Therefore, enhancing the reliability level of equipment and systems, from simple to complex and even highly complex networks, is of great importance. The concept of reliability was first introduced during World War II in the context of improving the dependability of military equipment. The metal powder production industry is exposed to the risk of dust explosions due to dust generation. Therefore, analyzing the production system, identifying critical equipment, and evaluating the reliability of the system are among the most important ways to reduce the probability of explosions. In this paper, the fuzzy reliability of an aluminum powder production system is investigated based on statistical data, and it is demonstrated that fuzzy reliability is significantly more accurate than classical reliability.

Abstract One of the common approaches in system reliability optimization is the use of redundant components. It has been proven that the redundant component allocation problem is NP-hard and involves selecting redundant components to optimize system reliability based on pre-defined constraints. In this paper, the maximization of reliability in series-parallel systems is addressed by adding redundant components subject to weight and cost constraints. In the allocation of redundant components, the existence of multiple types for each component is considered, meaning that in addition to determining the number of components, it is also necessary to select the appropriate type from the available options. This problem is modeled as a three-level graph, and an Ant Colony Optimization (ACO) algorithm is employed to solve it. The search capability of the proposed algorithm is enhanced by a local search method in the neighborhood of feasible points, and a dynamic penalty function is used to guide solutions toward feasible regions. The application of this algorithm is demonstrated in optimizing the reliability of a mechanical gearbox system. Numerical results obtained from solving sample problems indicate the considerable efficiency of the proposed algorithm compared to previous approaches, achieving not only the maximization of reliability but also minimizing the required weight and cost.




 

 




 

Abstract Sometimes, a quality characteristic is expressed as a specific functional relationship, which is referred to as a profile in statistical quality control. In this paper, linear profile analysis in Phase I control charts is investigated from a new perspective. In statistical quality control, it is usually assumed that the process has a constant mean. However, there may be processes where the mean of the parameters is not constant over time, and this variability is an inherent characteristic of the process. For controlling such processes, most existing control chart methods are ineffective. To statistically monitor such processes in Phase I, it is necessary to segment the process into sections where it is stable and consistent before performing statistical process control. For process segmentation, few studies have proposed the use of time-based clustering techniques, and the existing studies have mainly focused on univariate data. In this paper, we develop a process segmentation method for profile monitoring. Different segmentation algorithms for functional data are compared, and the results are reported. Studies show that the functional data clustering algorithm based on the Fuzzy c-means method performs well for profile segmentation.

Abstract Control charts are among the most important tools in statistical process control, used to monitor the extent of variation in a process. When a assignable cause deviation is observed in a control chart, identifying the root causes of the change and determining the time at which the deviation began—referred to as the change point—is crucial and impactful. In some statistical process control problems, the quality of a product or the performance of a process is described by the relationship between a response variable and one or more independent variables, known as a profile. In many applications, such as calibration, this relationship is described by a linear profile, while in other situations more complex models, such as Poisson regression profiles, are required. In this paper, the maximum likelihood estimation method is employed to detect change points in Phase II monitoring of Poisson regression profiles, and its performance is evaluated through simulation.




 

 




 

Abstract Statistical process control is a powerful method for establishing stability and improving process performance by reducing variability. Control charts are among the most widely used tools in statistical process control and play a significant role in enhancing process quality. Recent studies have shown that adaptive control charts—such as those with variable sample sizes, variable sampling intervals, and variable parameters—detect mean shifts faster than standard Shewhart xˉ\bar{x}xˉ control charts. One common assumption in designing a control chart is that observations follow a normal distribution; however, this assumption may not hold in some processes. In this paper, the performance of adaptive control charts under non-normal data is investigated using a simulation approach in MATLAB. It is demonstrated that the variable-parameter control chart outperforms other adaptive xˉ\bar{x}xˉ control charts in detecting small mean shifts under non-normal data, and, most importantly, it reduces the risk of false alarms.