Journal of Quality Engineering and Management

Abstract Determining the Factors Influencing the Prediction of Helicopter Rotor Failures

Purpose: The purpose of this paper is to investigate and identify the variables that influence the occurrence of helicopter accidents caused by different types of rotor failures. These crucial factors include flight conditions, maintenance conditions, and helicopter configuration. With this approach, accidents can be investigated more effectively and flight safety can be significantly improved.

Methodology: By analyzing 135 rotor faults accident from a comprehensive dataset containing 5652 helicopter-related accidents, eight classes of rotor faults were identified. Based on expert surveys and a review of studies in the field of helicopter accidents, nine features were proposed as crucial factors to such accidents. The significance of these factors was assessed using five feature selection methods. The input features included maximum takeoff weight, flight hours since the last inspection, type of last inspection, engine power, flight hours, altitude, wind speed, wind direction, and flight phase. Five well-known feature selection techniques—Correlation Matrix, Extreme Gradient Boosting (XGBoost), Mutual Information, Deep Learning, and Neural Network—were employed to identify the most essential factors.

Findings: "Maximum weight", "helicopter engine power", "flight phase" and "flight hours" were identified as variables with the highest degree of importance in predicting faults class of helicopter rotor, which also have a strong and acceptable justification in flight mechanics.

Originality/Value: The distinction of the present study from similar works lies in the inclusion of a broader range of variables, such as flight conditions and helicopter configuration, in contrast to previous studies that considered only a limited set of variables. By prioritizing these variables, the findings pave the way for proactive measures to prevent rotor faults, aiming to enhance prediction accuracy, reliability, and flight safety.

Abstract Purpose: The vaccine cold supply chain, as a quality-sensitive service–operational system, plays a critical role in ensuring timely delivery, maintaining vaccine efficacy, and minimizing wastage. The occurrence of operational and disruption risks intensifies process variability, undermines system reliability, and degrades service quality. The objective of this study is to develop a quality-oriented framework for the modeling and optimization of the COVID-19 vaccine cold supply chain, with a particular emphasis on quality-related performance indicators under conditions of uncertainty.

Methodology: In this study, a novel multi-period and multi-product simulation–optimization framework is developed to support decision-making in vaccine inventory management, allocation, and distribution under operational and disruption risks. The proposed approach integrates agent-based simulation with optimization techniques. The simulations are configured based on scenarios involving transportation disruptions and vaccine supply disruptions and are benchmarked against a disruption-free case.

Findings: The results are evaluated using several key performance indicators, including the expected vaccine delivery time, service level, vaccine wastage due to vehicle failures, and financial metrics. The simulation results indicate that the disruption-free scenario achieves the highest service level (0.82) and greatest degree of performance robustness, whereas transportation disruptions result in the spoilage of 8.9 million vaccine doses, and vaccine supply disruptions lead to the lowest service level (0.76). Statistical validation using the paired sign test further confirms the significance of these differences at the 95% confidence level.

Originality/Value: The present study adopts a quality engineering perspective to analyze the vaccine supply chain as a service system sensitive to process variability and proposes a quality-driven risk management framework. The findings provide practical insights for policymakers to enhance system reliability, reduce quality-related costs, and improve the resilience of vaccine distribution systems under crisis conditions.

Abstract Purpose: The purpose of this paper is to propose a novel integrated decision framework for risk-based maintenance strategy selection under uncertain and complex conditions. This paper aimed to address two critical research gaps: (i) inherent uncertainty in subjective expert judgments, and (ii) neglected interdependencies among risk evaluation criteria.

Methodology: The framework uniquely integrates three fuzzy-based methods: Fuzzy FMEA to identify and prioritize critical failure modes using linguistic variables; Fuzzy ANP to capture nonlinear interdependencies among risk factors and generate reliable criteria weights; and ELECTRE III to perform outranking-based ranking of alternative maintenance strategies under partial information conditions.

Findings: The framework is validated through a real-world case study of a Ballast Cleaning Machine in the Iranian railway network, using five years of CMMS historical data. Results demonstrate that Reliability-Centered Maintenance (RCM) significantly outperforms other strategies in mitigating high-risk failures and improving asset availability.

Originality/Value: The originality lies in the synergistic integration of FMEA, ANP, and ELECTRE III within a fuzzy environment—a combination not previously explored in railway asset management. This study offers a transparent, robust, and generalizable decision-support tool applicable to various complex industrial systems, including power plants, oil and gas, and manufacturing sectors.

Abstract The purpose of this paper is to evaluate the performance of classical survival analysis techniques and contemporary machine learning algorithms in predicting survival outcomes among patients diagnosed with heart failure. The study aims to determine whether machine learning–based models provide improved predictive accuracy compared to conventional statistical approaches. The study applied Kaplan–Meier, log-rank, Cox, and Weibull models alongside machine learning techniques including decision tree, random forest, IPCW logistic regression, SVM, and a survival neural network. Model performances were evaluated based on predictive accuracy. The Cox model performed best among classical methods, while IPCW logistic regression and random forest showed the highest accuracy among machine learning models. The survival neural network performed similarly to Cox and random forest without significant improvement. The paper provides a concise comparison showing that machine learning models can outperform traditional survival analysis approaches in predicting heart failure outcomes.

Abstract Purpose: Quality management in pharmaceutical supply chains requires the simultaneous evaluation of conflicting criteria, interdependencies among criteria, and uncertainty in decision-makers’ judgments. This study aims to develop an adaptive framework for pharmaceutical supplier selection that accounts for interdependencies among criteria and updates criterion weights based on performance feedback.

Methodology: This study proposes the DG-INFAO multi-criteria decision-making framework, which integrates graph-based aggregation with uncertainty-aware feedback learning. Within this framework, interval-valued trapezoidal neutrosophic numbers are used to represent uncertain evaluations, a graph structure is employed to model interdependencies among criteria, and an adaptive learning mechanism is used to update criterion weights. The framework is evaluated using simulated data for the selection of six pharmaceutical suppliers through convergence analysis, comparison with benchmark methods, ranking stability assessment, and sensitivity analysis.

Findings: The results indicate that the algorithm converges after 358 iterations, with the root mean square error (RMSE) decreasing from 0.067405 to 0.066898. The third and fifth suppliers rank first and second, respectively. Sensitivity analyses involving the graph aggregation parameter and random perturbations of the graph structure confirm the stability of these two suppliers’ rankings across the scenarios examined. However, cross-validation does not demonstrate the superiority of the proposed framework across all evaluation metrics. Therefore, further investigation is required before the findings can be generalized to real-world data.

Originality/Value: The main contribution of this study lies in integrating interval-valued neutrosophic evaluations, graph-based modeling of criterion interdependencies, and uncertainty-aware feedback learning within a unified framework. The framework’s added value is assessed through adaptive criterion-weight updating, ranking stability analysis, and sensitivity analysis of the results. Given the use of simulated data and the limited number of alternatives, the framework’s empirical validity and generalizability should be further examined in future studies using real-world data and larger decision-making sets.

Abstract Purpose: Linear autoregressive models are among the most widely used models in time series analysis. However, the presence of outliers and data with heavy-tailed distributions significantly reduces the accuracy of parameter estimation in these models. Therefore, the use of quantile models as an effective solution to address this challenge and improve analyses in fields such as statistical quality control is essential.
Methodology: In this article, the parameters of the Weighted Composite Quantile Autoregressive (WCQAR) model are estimated using a frequentist approach, and its application in statistical quality control is presented. The performance of the proposed method is evaluated through a comprehensive simulation study and the analysis of a real dataset (the percentage weight of a material in the output of a distillation column).
Findings: The results of the simulation and real data analysis show that the SEM algorithm, compared to conventional methods, provides more accurate and stable estimates and effectively overcomes problems caused by outliers and heavy-tailed distributions.
Originality/Value: The main innovation of this research is the use of the SEM algorithm instead of conventional algorithms such as EM to compute the maximum likelihood estimators for the parameters of the weighted mixture quantile autoregressive model. This approach can serve as a powerful and reliable tool in time series analysis and statistical quality control.

Abstract Objective: The National Quality Assurance Framework (NQAF), developed by the United Nations, provides a systematic framework for assessing and improving the quality of official statistics across four areas: statistical system management, institutional environment management, statistical process management, and statistical output management. Given the strategic role of official life insurance statistics in economic, social, and welfare analyses, and impact on policymaking and industry regulation, the quality of statistical reports is vital. However, in Iran’s insurance industry, no context-specific and coherent framework has yet been developed for assessing the quality of such reports based on international standards. Accordingly, this study aims to identify quality challenges in official life insurance statistical reports and develop a context-specific framework for their improvement based on NQAF principles.

Methodology: This study is applied and descriptive-analytical, using an exploratory mixed-methods approach in four stages. In the first stage, official statistical reports and documents concerning life insurance were reviewed. In the second stage, quality challenges were identified through expert focus groups and analyzed using thematic analysis. The third stage involved mapping the identified challenges against the principles of the NQAF and designing localized indicators. Finally, in the fourth stage, the proposed checklist was validated through expert evaluation, practical testing on statistical yearbooks, and a pilot implementation within an insurance company.

Findings: Major quality challenges include insufficient disaggregation of certain statistical items, potential discrepancies between yearbook data and financial statements, inadequate standardization of definitions and metadata, the absence of systematic mechanisms for validating input data, limited transparency and accessibility of information, and weaknesses in online reporting infrastructure. Accordingly, a set of indicators and improvement actions was developed in two areas: statistical system management and statistical output management. Contextualizing the NQAF framework for life insurance statistics can provide an operational basis for improving the accuracy, consistency, transparency, reliability, and efficiency of official statistics. The proposed checklist can serve as a practical tool for regulatory authorities, statistical data producers, and policymakers, improving decision-making and the insurance industry’s statistical system.

Originality/Scientific Value: Accordingly, the added value of the present study lies in moving from the general level of the NQAF to the specific domain of life insurance, in designing and preliminarily validating an operational 45-indicator checklist, and in linking the specialized and regulatory requirements of life insurance to statistical quality principles. This framework can provide a basis for the systematic assessment of the quality of life insurance statistical reports in Iran.

Abstract Purpose: This study aims to develop a scenario-based multi-objective decision support framework for upgrading the value chain of perishable agricultural products. The proposed framework incorporates the initial post-harvest quality of products to enable appropriate allocation among different pathways, including fresh consumption, processing, and value recovery from low-quality products and waste through secondary product generation. This approach contributes to reducing product losses, improving waste management, enhancing the economic value of the value chain, and promoting resource sustainability.
Methodology: A multi-objective, multi-period, and scenario-based optimization model was developed for the agricultural value chain, incorporating production and demand uncertainties as well as transportation budget constraints. The model objectives include maximizing total profit while minimizing the water footprint and carbon emissions. The proposed model was solved using the ε-constraint method in GAMS, and its validity and robustness were evaluated through sensitivity analysis.
Findings: The results obtained from the case study of the tomato product value chain in Mazandaran Province, based on market data and expert judgments, demonstrated that the proposed model is capable of achieving a balance between economic and environmental objectives under various uncertainty conditions. The baseline scenario results confirmed the profitability of the value chain and the dominance of revenues over costs. Sensitivity analysis revealed that the prices of fresh and processed products have the greatest impact on economic performance, while processing capacity is not the primary limiting factor of the system. Furthermore, the investigation of the processing conversion parameter indicated that increasing yield can improve economic performance only when sufficient demand exists for processed products.
Originality/Value: The main contribution of this study lies in developing a Scenario-based multi-objective decision support framework that integrates quality-based product allocation, waste management and recovery, and sustainability objectives within a unified framework. This framework facilitates coordinated decision-making to reduce product losses, enhance value creation, and improve the overall sustainability of the agricultural value chain.

Abstract Purpose: Traditional banking needs new FinTech innovations and technologies to improve its processes and services. FinTech innovations have led to significant changes in the banking system, including advancements in risk management. Therefore, the present study aimed to investigate and analyze the heterogeneity and the mechanism underlying the effect of FinTech innovation on the risk-taking of commercial banks using balanced panel data from 20 banks for the period 2013-2022.
Methodology: Based on web technology, an indicator at the bank level is considered, including the creation, annual number, and frequency of news related to fintech innovation from each bank. This indicator is calculated as the ratio of the value of online shopping and bill payments made through the Internet and mobile devices to GDP. To address potential endogeneity issues, including measurement errors and omitted variables, the methods of Instrumental Variables (IV) and Difference-in-Differences (DID) were employed to test the hypothesis and obtain consistent estimates.
Findings: Showed that improvement in FinTech bank innovation significantly reduces risk-taking. The results of the mechanism analysis indicate that a bank's FinTech innovation reduces its risk-taking through two channels: increasing operating income and enhancing the capital adequacy ratio. The analysis of the heterogeneity of bank size, bank type, and competitiveness shows that larger, public, private, and highly competitive commercial banks have a more pronounced effect on reducing risk-taking in the development of technological innovation. Also, robustness and stability tests, including changing the methods used to construct the FinTech innovation index, replacing risk-taking indicators, and reducing the change in the study sample, showed that the findings remained unchanged.
Originality/Value: The banking system should adopt a development model aligned with the era and utilize FinTech solutions to accelerate its digital transformation. Finally, since the use of FinTech by commercial banks presents certain potential risks, banks should enhance their risk management. Implement applicable supervisory measures, such as information disclosure standards and risk management indicators.

Abstract Purpose: The existing supervisory system in many countries, especially in Iran, mainly uses general and integrated models in which the structural, mission, and operational differences between banks and financial institutions have not been properly taken into account. This uniform approach has led to reduced risk identification accuracy, a lack of adaptation to each financial institution's specific needs, and reduced effectiveness of supervisory measures. The main objective of the present study is to design a cause-and-effect model to improve the quality of supervision based on the mission type of banks and financial institutions, using a mixed approach (Meta-Synthesis-Fuzzy DEMATEL).
Methodology: The present research was conducted using a mixed method (Qualitative-quantitative) and exploratory approach. Then, using a survey, 25 banking industry experts with at least 10 years of executive experience in finance and banking and master's and doctoral degrees were recruited to examine the validity and reliability of the proposed model. Also, paired-comparison questionnaires were distributed to the experts, and the intensity of impact and effectiveness between the research dimensions were examined using the fuzzy multi-criteria decision-making technique, DEMATEL.
Findings: The research findings show that using the meta-synthesis approach, 9 dimensions and 40 components were selected. They were selected from the dimensions. Also, the results of fuzzy DEMATL analysis show that the most influential dimension of the present study regarding the supervision of banks and credit institutions based on the value of (D+R) is the legal and regulatory supervision dimension among the independent dimensions and the cause with the highest value and the most influential variable. Also, among the dependent and affected dimensions based on the lowest value (D-R), the environmental and social performance dimension was recognized as the most influential variable in improving the quality of supervision of banks and credit institutions.
Originality/Value: Using a meta-combination approach in the analysis of previous research, which can provide new horizons for designing effective models based on improving the quality of banking system supervision. Therefore, the present study is of high scientific and practical importance for advancing methodology, responding to the current needs of the country's financial system, and improving the effectiveness of supervision of monetary institutions. This research has also led to recognition of the intensity of the relationships among the dimensions of quality improvement in banking industry supervision and has drawn more attention to these components.

Abstract Purpose: This study aims to evaluate and rank collaboration models in the Iranian rubber industry supply chain from the perspective of digitalization quality. Digitalization quality refers to the effective use of Industry 4.0 technologies to improve transparency, integration, agility, resilience, and sustainability. The rubber industry was selected due to its operational complexity and urgent need for digital transformation.
Methodology: A multi-criteria decision-making approach was adopted, combining the Fuzzy Best-Worst Method (BWM) for weighting the evaluation criteria and TOPSIS for ranking the collaboration models. A sensitivity analysis was also conducted to assess the robustness of the results across varying criterion weights.
Findings: The digital supply chain model ranks highest in digitalization quality, with "technology integration" as the most critical criterion. The sensitivity analysis confirms the rankings' robustness and stability across different weight scenarios.
Originality/Value: This research uniquely addresses the comparative assessment of collaboration models in the rubber industry based on digitalization quality. The use of a Fuzzy BWM-TOPSIS hybrid method and comprehensive sensitivity analysis provides a novel, practical framework for strategic decision-making in digital supply chain transformation.

Abstract Purpose: This work addresses the automotive industry with two important, evolving concepts: sustainability and resiliency. The proposed model is designed to balance economic, environmental, and social objectives while maintaining adaptability to potential supply chain changes and disruptions. A real automotive company is then investigated as the case study to assess the applicability, validity, and performance of the developed model and, eventually, render useful managerial and decision aids.
Methodology: To achieve this objective, a comprehensive decision-making model has been developed. In the first stage, supplier evaluation is conducted based on sustainability and resilience criteria. This assessment employs two innovative decision-making approaches: the stochastic fuzzy Best-Worst Method (BWM) and stochastic VIKOR. In the subsequent stage, a multi-objective mathematical model is formulated by incorporating stochastic-fuzzy uncertainty. To solve the model, a fuzzy robust optimization approach combined with a modified multi-choice goal programming method based on a utility function is applied.
Findings: In this study, the supply chain of SAIPA Kashan Automotive Company is analyzed across three key dimensions: general criteria, sustainability, and resilience. Indicators such as cost, quality, and reductions in energy consumption were identified as the most critical evaluation factors. Supplier evaluation and ranking were carried out using the fuzzy VIKOR method. The results indicate that, among the main suppliers, the second and fifth options, and among the backup suppliers, the second option, received the highest scores. Furthermore, to assess the robustness and validity of the proposed approach, the results were compared with those obtained using conventional methods.
Originality/Value: The value of this research lies in presenting a comprehensive decision-making model under uncertainty and in improving supply chain performance across economic, environmental, and social perspectives. The findings can significantly assist managers and policymakers in the automotive industry in addressing complex supply chain challenges.

Abstract Purpose: This research aims to identify and examine the relationship between the barriers to implementing Quality 4.0 in the Yazd tile and ceramic sector.
Methodology: A systematic review was employed to identify and categorize barriers, while the combined Fuzzy DEMATEL-ANP (FDANP) method was used in the quantitative phase to elucidate the relationships among these barriers and to prioritize them.
Findings: The qualitative analysis revealed 18 barriers across four categories. The Quantitative findings indicated that "High Cost of Implementation and lack of transparency in Return on Investment" emerged as the most causal barrier, while "failure to consider Quality 4.0 as a strategic issue and source of competitive advantage" was identified as the most consequential barrier to implementing Quality 4.0 in the Yazd tile and ceramic industry. Also, the key obstacles to the implementation of Quality 4.0 in the Yazd ceramic and tile industry include the lack of quantitative metrics for assessing the impact of Industry 4.0 on quality, the failure to consider Quality 4.0 as a strategic issue and a source of competitive advantage, the absence of advanced training programs for personnel, insufficient financial resources for implementing Quality 4.0, and the lack of stakeholder involvement in Quality 4.0 initiatives and projects.
Originality/Value: This study employs a hybrid Fuzzy DANP approach to support managers and decision-makers in strategizing for overcoming obstacles to the adoption of Quality 4.0. It achieves this by identifying barriers, determining their interrelationships, and prioritizing them.

Abstract Purpose: This study aims to determine the optimal stock portfolio using a combination of the Markowitz model with Support Vector Machine (SVM), Data Envelopment Analysis (DEA), and the DBSCAN clustering algorithm. The statistical population consists of companies listed on the Tehran Stock Exchange from 2012 to 2022.
Methodology: To achieve the research objectives and form an optimal stock portfolio, dimensionality reduction approaches, DEA, SVM, and the DBSCAN clustering algorithm were employed. Financial ratios derived from balance sheets, income statements, and cash flow statements, as well as composite financial ratios and risk-return analysis based on the hybrid Markowitz model, were used as inputs to construct four portfolios.
Findings: The SVM method and the fourth approach, which includes the hybrid model, exhibited superior performance in optimizing the stock portfolio.
Originality/Value: Given the innovation of this research in applying the hybrid Markowitz model, the results can assist investors and stock analysts in managing the quality of an optimal stock portfolio.

Abstract Purpose: In light of the growing challenges in today's supply chains, including market fluctuations, increasing environmental and social pressures, and the need to enhance resilience against foreseeable crises such as the COVID-19 pandemic and economic disruptions, the strategic importance of selecting suppliers that simultaneously meet sustainability and resilience criteria has become more prominent. Accordingly, the main objective of this study is to present a comprehensive, data-driven, and forward-looking decision-making model for evaluating and selecting suppliers within the supply chain, accounting for multiple dimensions of sustainability and resilience simultaneously.
Methodology: In the proposed model, the weights of the defined criteria and sub-criteria were initially determined using the Stochastic Best-Worst Method (SBWM). Supplier performance was then evaluated using the Stochastic VIKOR Multi-Criteria Decision-Making (MCDM) method. In the final stage, the Random Forest regression algorithm was applied to predict future supplier performance. The model was tested through a case study conducted at SAIPA Kashan Automotive Company using expert input collected via structured questionnaires.
Findings: Sustainability and resilience criteria play a central role in supplier selection in the automotive industry. Among the sub-criteria, "greenhouse gas emissions" and "energy consumption reduction" were most influential due to environmental regulations. At the same time, "cost" and "safety stock level" had the greatest impact due to their direct effect on economic performance and operational continuity. Furthermore, the Random Forest algorithm achieved high predictive accuracy (RMSE = 0.0976), confirming the model's ability to generate reliable, data-driven forecasts.
Originality/Value: Although each of the methods used in this research (Random Best-Worst Method, Random VIKOR, and Random Forest algorithm) has been employed individually in previous studies, the main innovation of this study lies in presenting an integrated framework that combines all three approaches. In fact, this research is the first to merge MCDM methods with a machine learning algorithm, offering a comprehensive, data-driven decision-making model. This model not only assesses the current performance of suppliers but also enables prediction of their future performance. Such a combination has not previously been introduced in the supplier selection literature with a simultaneous focus on supply chain sustainability and resilience in the automotive industry, marking a clear methodological innovation.

Abstract Purpose: In today's competitive environment, selecting suitable suppliers plays a pivotal role in enhancing the efficiency and sustainability of supply chains. The LARG supply chain model, as a comprehensive approach, integrates various dimensions in supplier management. However, despite its critical influence on interaction success and supplier selection, the cultural dimension has received limited attention. This study aims to evaluate suppliers within the LARG framework while incorporating the cultural dimension to improve supply chain performance and achieve sustainable competitive advantage.
Methodology: This research developed a data-driven and forward-looking decision-making model for supplier evaluation and selection. First, key criteria were identified through literature review and expert consultation, and weighted using the Fuzzy Best–Worst Method (FBWM). Then, suppliers' efficiency was assessed and ranked using Fuzzy Data Envelopment Analysis (FDEA). Subsequently, the Random Forest algorithm was employed to predict future supplier performance, yielding highly accurate results.
Findings: The initial results highlighted the significance of sub-criteria such as greenhouse gas emission reduction, risk management, quality, and delivery speed in supplier evaluation. In the second phase, supplier efficiency was analyzed under various α-cuts and classified into three performance groups: high, medium, and low. The Random Forest model demonstrated high accuracy in forecasting supplier performance. Moreover, the paired t-test results revealed that incorporating the cultural dimension significantly improves the supplier selection process.
Originality/Value: The proposed model contributes to strategic decision-making by identifying key performance factors, enabling predictive evaluation, and employing robust analytical tools. This approach not only reduces risks and costs but also serves as a practical model for improving supply chain performance in similar industries.