نشریه مهندسی و مدیریت کیفیت
سال یازدهم شماره 4 (زمستان 1400)

In shock models, the system will fail as soon as the amount of damages caused by shocks exceeds its failure threshold. In this research, the design of control chart for reliability monitoring of systems with random failure threshold subjected to cumulative random shocks is discussed. In the cumulative shock model, the system fails when the cumulative amount of damage caused by the shocks exceeds the failure threshold of the system. For the analytical solution of the model related to the control chart, there are some complexities such as the existence of unknown distributions, obtaining the convolutions of the distributions and integral calculations. To overcome these problems, phase-type distribution is used for modeling. By presenting a numerical example, the results of the presented analytical model are evaluated and compared with the Monte Carlo simulation method. Also, the performance of the proposed control chart using the average run length and average time to out of control signal criteria is evaluated.

Human reliability is an important issue that is evaluated in order to prevent catastrophic events in high technology industries. The purpose of this study was to evaluate human reliability by CREAM fuzzy method in short and medium range radar systems, which ultimately led to the study of human reliability and the probability of error in each of the effective factors. Then the method of hierarchical analysis, pairwise comparisons, calculation of weights and system compatibility, matrix of pairwise comparisons and control style, and ranking were performed. After collecting information from questionnaires and consulting experts, job analysis was performed by hierarchical analysis (HTA). Then the table of working conditions affecting the operator's performance (CPC) was drawn, and working conditions affecting the user's performance were presented using the table of CPCs. In the next step, the fuzzy AHP method was investigated , after calculating the pairwise comparison matrices, control styles were determined. At the end of the work, the factors causing the error along with the amount of human reliability and the probability of error in each factor were identified. The results indicated that human reliability is significantly improved. Based on the findings of the study, determining the contribution of various factors influencing human error, provide a more effective assessment. Among the job tasks, the factor of "preparation of mechanical parts" with a probability of 0.416 has the highest probability of error and should focus more on this factor.

Currently, rapid economic change and increasing competitive market pressure are pushing organizations to focus on making supply chain operations more efficient and effective. Proper design and efficiency of logistics networks as part of supply chain planning, in addition to creating a sustainable competitive advantage, increases customer satisfaction and provides the opportunity to meet their needs, which is why the decisions related to the design of these networks are of great importance. Enjoy. Therefore, in this study, the design of a closed-loop logistics network to reduce pollution and environmental pollution using the Bertsimas and wire stabilization method was presented. The mathematical model to be presented in this research was presented by considering the objectives of minimizing transportation costs, minimizing the time of receiving raw materials from the supplier and minimizing the time of product return from the customer to the separation center. Due to the strategic nature of the closed-loop supply chain, which with the approximate solution space causes a lot of costs to be delivered to the system to increase the accuracy of the answers of the mathematical model and application of this goal in this study It is used to reduce the computational time of the model, the results obtained with high accuracy. On the other hand, because the operational logic of solving Lagrange release is based on a single-objective model, first multi-objective mathematical model with Augmented Epsilon-Constraint The target was converted and then the Lagrange release algorithm was implemented on it.

Nowadays, statistical analysis and monitoring of networks and early detection of anomalies with a significant growth rate have received more attention than before in recent years. In the real world, there is a wide range of networks analyzed and improved through network monitoring solutions, such as transportation, supply-demand, financial exchanges, health care, as well as the social ones, the analysis of the results can be beneficial to the stakeholders. The basis of the research is on identifying and solving the real problem. In other words, a real problem is identified in the country and a methodology is developed to solve it. The case study is the monitoring of a network of centers that provide emergency services in cities. The nature of this network is dynamic, feature-based, directed and weighted. The results of this study show that by modeling complex systems as a network and its continuous monitoring, abnormal situations can be identified and managed early and crises in cities can be prevented.

As time goes on and crises increase in societies, organizations are increasingly exposed to disruption. These crises can be of natural (such as earthquakes, floods, and fires) or human (such as terrorist attacks, infectious diseases, and intentional or inadvertent employee errors). Therefore, organizations need to be resilient to protect themselves from harmful consequences. The basic aspect of resilience involves the ability of an element to return to normal after disruption and resource allocation. Obviously, in any organization, the primary goal is to allocate the least resources to recover operations and to bring activities back to the tolerance threshold so that destructive events do not stop vital activities. In this paper, a quantitative model for resource allocation is presented, which minimizes the lack of resilience. The problem has a basic assumption, that there is a shortage of resources in at least one of the available resources due to excessive demand. After solving the model by numerical experiment, the results of the model were described and it was found that destructive events were retrieved before the tolerance threshold.

Recent years, the insurance industry has been experiencing an increase in equipping insurance companies with fraud detection systems. Furthermore due to the significant cost imposed on the insurance industry by the rise in such claims, the role of data mining techniques in detecting fraudulent claims has become widespread. However an essential issue with such systems is the quality of their outputs. On one hand, supervised algorithms are more accurate comparing to unsupervised counterparts. On the other hand, as data labeled fraud is really limited, the efficiency of supervised algorithms is severely challenged. Within this regard, a novel approach is introduced as “alternative feature” to overcome the challenge. Basically, alternative feature is a variable whose values are available and can be considered a suitable indicator to detect suspicious cases. This approach improves the efficiency of the system and allows experts and insurance companies to investigate suspicious cases with more confidence and less error.