University of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701A Hybrid Approach Based on Data Density Analysis for Group Supplier Selection Problem Considering Multi-Level Risk Assessment: A Case Study130482310.22091/jdaid.2026.16147.1066ENMohammad RezaMarjaniDepartment of Industrial Engineering , Qom University of Technology, Qom, Iran0000-0002-7781-2651MohammadHabibiDepartment of Industrial Engineering, Qom University of Technology, Qom, Iran0009000619601246MehdiJabbariDepartment of Computer Engineering, Qom University of Technology, Qom, Iran0009-0008-7674-4408MajidAghaeeMSc, Faculty of Electrical and Computer Engineering, Qom University of Technology, Qom, Iran. Email: aghaee@qut.ac.irJournal Article20260527Abstract: Supplier selection can be considered as the most important activity in the purchase process because it significantly affects the organization. Despite the presence of powerful suppliers, a variety of risks can be thought of in front of organizations. Thus, it is necessary to identify the related risks, and important decisions should be made by the organization concerning the suppliers with different levels of risks. The problem under consideration, is the group Suppliers selection based on different levels of related risk Criteria. The multi-stage method starts with identifying the risk and after determining the final risk Criteria, the weights of the Criteria are determined using the BWM method. Weighted matrix based on identified quantitative and qualitative risks commensurate with suppliers information and weights of the Criteria, will be used as input to the DBSCAN clustering process. DBSCAN is an unsupervised approach and used here for the supplier selection problem for the first time. We considered the Mega Motor Company, which is the largest manufacturer of powertrains in the Middle East and deals with hundreds of domestic and foreign suppliers. Clustering was made for different values of ε and Minpts. The inter- and intra-cluster distances were used to evaluate the results and clustering quality. Then, the k-means method and unsupervised machine learning were used to assess the validity of the proposed approach. A comparison of inter- and intra-cluster distances of the two approaches verified the efficiency of the DBSCAN approach in dealing with the supplier selection problem.https://jdaid.qom.ac.ir/article_4823_5540e005031b6acb327db8404164219b.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701From Analytics to Agility: How Smart Contract Capability and Supply Chain Digitalization Strengthen Supply Chain Performance in Financial Services3161482410.22091/jdaid.2026.16727.1094ENHosseinBahadori JahromiPh.D. Candidate in Financial Law, Faculty of Economics and Financial Sciences, Islamic Azad University, Science and Research Branch, Tehran, IranJournal Article20260501The growing digitalization of financial services has transformed supply chain operations, creating new opportunities to improve responsiveness and organizational performance through advanced data-driven and automation technologies. Despite substantial investments in big data analytics, many financial service organizations continue to struggle to convert analytical insights into agile operational capabilities and sustainable performance improvements. This study develops and empirically tests an integrated framework that explains how Big Data Analytics Capabilities (BAC) contribute to Supply Chain Performance (SCP) through the mediating role of Supply Chain Agility (SCA), while examining the contingent effects of Supply Chain Digitalization (SCD) and Smart Contract Capability (SCC).<br />Data were collected through a structured survey administered to managers from financial institutions, digital banking providers, fintech companies, electronic payment service providers, and financial technology firms. Hierarchical regression analysis and PROCESS bootstrapping techniques were employed to examine the proposed direct, mediating, and moderating relationships.<br />The findings demonstrate that BAC significantly enhances SCA, which subsequently improves SCP, confirming the mediating role of agility. In addition, both SCD and SCC strengthen the positive influence of BAC on SCA as well as the impact of SCA on SCP. These results indicate that organizations with more advanced digital infrastructures and stronger smart contract capabilities are more successful in transforming analytical intelligence into agile operational responses and superior supply chain outcomes.https://jdaid.qom.ac.ir/article_4824_41e2a6207d612d41ea65e2c4a3fee200.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701A Distributed Fuzzy Expert System on Edge-Cloud Architecture for Technology Commercialization Success Prediction6281482510.22091/jdaid.2026.16404.1077ENJafarYazdimoghaddamCorresponding author, Assistant professor, Department of Industrial Engineering, Faculty of Modern Technologies, Hakim Sabzevari University, Sabzevar, Iran. Email: j.yazdi@hsu.ac.irMortezaMohammadiAssistant professor, Department of Economics, Faculty of Letters and Humanities, Hakim Sabzevari University, Sabzevar, Iran. Email: drmohammadi1358@gmail.comJournal Article20260615Accurately predicting the commercialization potential of emerging technologies is essential for guiding investment decisions and shaping innovation policies. However, most existing fuzzy expert systems are based on centralized architectures, which limit their scalability when applied to large technology portfolios in open innovation environments. This study introduces a distributed fuzzy inference system (D-FIS) implemented within an edge–cloud computing framework to overcome these limitations. The proposed model evaluates technology commercialization using 32 indicators covering technological, financial, market, and regulatory dimensions. These indicators were identified through the fuzzy Delphi method, while decision-making is performed using Mamdani fuzzy inference and centroid defuzzification. To improve computational efficiency, rule evaluation is distributed across edge nodes, whereas the cloud layer is responsible for aggregating inference results and maintaining synchronization of membership functions. The framework was validated using a real-world dataset comprising technology commercialization cases. Experimental results showed an overall prediction accuracy of 72.73%, with a specificity of 94.44% and a sensitivity of 75.00%. Compared with a conventional centralized implementation, the proposed approach reduced inference latency by 64% while demonstrating near-linear scalability across up to 32 computing nodes. These findings indicate that the proposed D-FIS offers an interpretable and scalable decision-support framework for assessing commercialization risks under uncertainty and can facilitate collaborative decision-making in open innovation ecosystems.https://jdaid.qom.ac.ir/article_4825_304379d3e9711b5e5ba5130a1ada65cd.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701From Prediction to Explanation: Earnings Management Analysis Using Explainable Machine Learning and the SHAP Approach82121482610.22091/jdaid.2026.16254.1070ENSaeedAlipourDepartment of Accounting, Ard.C., Islamic Azad University, Ardabil, Iran0000-0002-1802-4460MehdiFazliDepartment of Mathematics, Ard.C., Islamic Azad University, Ardabil, Iran0000-0002-4089-1827Journal Article20260605Objective: This study aims to develop and evaluate an explainable machine learning framework for detecting earnings management and to compare its predictive performance with traditional accrual-based approaches in the Iranian capital market. Beyond improving prediction accuracy, the study explains the nonlinear relationships between financial, corporate governance, and market variables associated with earnings management using explainable artificial intelligence techniques.<br /><br />Method: This applied study adopts a descriptive-analytical design based on machine learning and explainable artificial intelligence. The statistical population consists of firms listed on the Tehran Stock Exchange during 2016–2025. After applying the sample selection criteria, 1,620 firm-year observations were analyzed. Five machine learning algorithms—XGBoost, Random Forest, Support Vector Machine, Decision Tree, and Logistic Regression—were evaluated and compared with a traditional accrual-based approach. Model interpretation was performed using SHAP (SHapley Additive exPlanations), while predictive performance was assessed using Accuracy, Precision, Recall, F1-score, and AUC.<br /><br />Results: XGBoost achieved the highest predictive performance, with an AUC of 0.90 and an F1-score of 0.86, significantly outperforming the traditional accrual-based approach and the other machine learning models. SHAP analysis revealed that operating cash flow volatility, financial leverage, market-to-book ratio, audit quality, and other financial and governance variables exert substantial nonlinear effects on earnings management prediction.<br /><br />Conclusions: The findings demonstrate that explainable machine learning not only improves predictive performance but also enhances model transparency and theoretical interpretation. The proposed framework provides valuable analytical support for auditors, regulators, investors, and corporate decision-makers seeking to strengthen financial reporting quality, regulatory oversight, and earnings management monitoring.https://jdaid.qom.ac.ir/article_4826_912ea27d67e564630be31505830a97dd.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Artificial Intelligence as a Catalyst for Strengthening the Strategic Value of Quality Management122141482810.22091/jdaid.2026.15903.1049ENMostafaTamtajiAssist. Prof., Dept. of Strategic Management, Faculty of Management and Industrial Engineering, Malek Ashtar University of Technology, Tehran, Iran.0000-0002-5825-6258AlirezaEkrami KivajPh.D., Dept. of Aerospace Engineering, Faculty of Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran0000-0003-4805-5046Sayed Gholam HassanTabatabaeiAssistant Professor of the Faculty of Electrical and Computer Engineering, Malek-e-Ashtar University of Technology, Tehran, IranJournal Article20260508This study proposes a comprehensive vision for integrating Artificial Intelligence (AI) into quality management practices, emphasizing its strategic contributions to organizational performance, process reliability, and continuous improvement. To evaluate this impact, we conducted a systematic review of recent literature combined with a simulation-based evaluation. This research specifically explores how AI technologies enhance critical quality functions, such as automated inspection, predictive auditing, and real-time performance monitoring.Our results, supported by regression and correlation analyses, indicate a significant positive association between AI implementation and key performance indicators, specifically in defect reduction, process optimization, and adherence to quality standards. However, the effective adoption of AI also presents notable hurdles. Challenges regarding data quality and integrity, algorithmic transparency, and the need for greater organizational readiness remain critical considerations. Addressing these factors is essential for ensuring reliable and responsible implementation. Ultimately, this study introduces a strategic framework designed to assist organizations in adopting AI-driven solutions. This approach allows businesses to harness the power of intelligent systems while maintaining ethical principles, ensuring that their quality management practices remain sustainable and effective in an increasingly digital environment.https://jdaid.qom.ac.ir/article_4828_da95390d2a25b387cb37d874f2b4d734.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Design of an Intelligent System for Optimal Discount Allocation in E-commerce Based on Customer Purchase Behavior142153482910.22091/jdaid.2026.16144.1071ENAliSarabadaniPhD in IT Engineering, Department of Information Technology Engineering
(E-commerce) ,Qom university, Qom, Iran.0000-0002-3521-1401MahtabTajvidianDepartment of Information Technology Engineering (E-commerce) ,Qom university, Qom, Iran.FatemehSadeghiDepartment of Statistics, Faculty of Statistics, Mathematics and Computer Science, Allameh Tabataba'i University, Tehran, Iran.MehrdadSaffarieDepartment of Information Technology Engineering (E-commerce),
K.N. Toosi University of Technology, Tehran, Iran.Journal Article20260606This study presents an intelligent, data-driven framework to optimize discount allocation strategies within e-commerce environments. Rather than deploying standard uniform promotional tactics, this research proposes a prescriptive model that segments customers based on transaction history to maximize aggregate profitability and mitigate high-value customer attrition.Utilizing an empirical transaction dataset containing 4,570 validated purchase records from 1,000 unique customers, we calculate Recency, Frequency, and Monetary (RFM) vectors. Following min-max normalization, K-means clustering partitions the customer base. The optimal cluster structure (K=4) is formally validated using the Silhouette Coefficient (S) and the Davies-Bouldin Index (DB) alongside the traditional elbow method. We then formulate a constrained non-linear optimization model utilizing a logistic customer response function to analytically derive segment-specific discount rates that maximize expected net profit under a fixed marketing budget constraint. The empirical system identified four distinct customer archetypes: loyal high-value, at-risk, recent low-frequency, and inactive low-value customers. Compared to a baseline uniform 10 percent discount policy, the proposed optimized framework increased total profit from $184,965.74 to $197,556.92, representing an 8.7 percent improvement. Net profit margins expanded from 21.8 percent to 23.4 percent.Concurrently, high-value customer churn decreased by 12.3 percentage points, a reduction proven statistically significant via McNemar's test (p < 0.001), while promotional waste in low-value segments was entirely eliminated.These findings demonstrate that integrating behavioral clustering with constrained optimization structures yields significantly higher economic returns and customer retention compared to uniform pricing approaches. The proposed system is highly interpretable, computationally efficient, and directly applicable for deployment within small-to-medium enterprise (SME) e-commerce platforms.https://jdaid.qom.ac.ir/article_4829_3eecdc6152799bc2d1b5a84ae4095733.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Predicting Remaining Time in Object-Centric Process Executions Using Machine Learning Models154166487710.22091/jdaid.2026.16354.1075ENMohsenMohammadiCorresponding Author, Department of Computer engineering, Esfarayen University of Technology, Esfarayen, Iran. Email: mohsen@esfarayen.ac.ir0000-0003-0532-1236Journal Article20260612Objective: Predictive process monitoring is critical for effective business process management; however, current approaches relying on case-centric logs often fail to capture the dynamics of complex, multi-object interactions. This study proposes a framework for object-centric remaining time prediction, facilitating proactive operational decisions, including systematic delay reduction, optimized resource allocation, and enhanced service quality.<br />Method: We developed a prediction methodology leveraging object-centric event data from real-world logs. Our feature engineering captured multidimensional information, including temporal dynamics, elapsed execution time, sequential activity history, and intricate interactions between interrelated process objects. These features were used to train and compare three machine learning architectures: a Linear Regression baseline, Random Forest, and the Gradient Boosting algorithm.<br />Results: Empirical analysis reveals that ensemble tree-based methods consistently outperform the linear baseline. Specifically, the Gradient Boosting model demonstrated superior prediction accuracy, successfully mitigating the noise inherent in complex process data. Feature importance analysis confirmed that elapsed time, activity histories, and object-centric interaction features serve as the most influential predictors for remaining time.<br />Conclusion: This research confirms that synthesizing object-centric feature extraction with ensemble learning provides a practical solution for time prediction in complex, multi-object environments. By addressing the limitations of case-centric assumptions, this approach enables more reliable forecasting and enhances operational decision-making in diverse real-world process management contexts, thereby bridging a significant research gap.https://jdaid.qom.ac.ir/article_4877_f9332a2faa1d6e1d5a1498e53a5dbca0.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701A Hybrid CNN-HMM Framework for EEG-Based Classification of Anxiety, Chronic Stress, and Depression167184487610.22091/jdaid.2026.16603.1082ENAmir HosseinKahvandDepartment of Electrical and Computer Engineering, Hamedan University of Technology, Hamedan, Iran0009-0000-5816-5164Hamid RezaShahdoostiDepartment of Electrical and computer Engineering, Hamedan University of Technology, Hamedan, Iran.MohammadGhahremaniDepartment of Electrical and Computer Engineering, Hamedan University of Technology, Hamedan, IranJournal Article20260106One of the biggest challenges in clinical medicine is early and accurate diagnosis of brain disorders, especially functional and psychological disorders like stress, anxiety and depression. While conventional machine learning relies on hand engineered time-consuming features, deep learning models, despite showing superior accuracy, can be limited in clinic application by interpretability issues because of their “black-box” design. In this paper we present a novel hybrid approach for brain disorder detection and classification through the analysis of EEG signals. Our model combines the strength of automatic feature extraction of Convolutional Neural Network (CNN) and time sequence modelling & interpretation of Hidden Markov Model (HMM). A real clinical data from 200 individuals (healthy group, anxiety group, chronic stress group, depression group) is collected and used to train and evaluate our system. This local collected data is for the first time used to validate models. Our results show that the hybrid approach achieved 92.30 percent in 4-class classification, and showed superiority to other standard approaches. The interpretable nature by HMM transition matrices also make it more relevant to aid clinical decision-making.https://jdaid.qom.ac.ir/article_4876_7a326d6cc16f8b5b7de1fff832a61d51.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Robust Mathematical Modeling for Air Multiple Logit Design in Flight Route Selection185211487810.22091/jdaid.2026.16639.1084ENMohammad AliKamali BandpyDepartment of Industrial Engineering, AK.C., Islamic Azad University, Aliabad Katoul, IranHosseinAmoozad KhaliliDepartment of Industrial Engineering, Sar.C., Islamic Azad University, Sari, Iran0000-0001-7222-2233MohammadAmirkhanDepartment of Industrial Engineering, AK.C., Islamic Azad University, Aliabad Katoul, IranAhmadMehrabianDepartment of Industrial Engineering, AK.C., Islamic Azad University, Aliabad Katoul, IranJournal Article20260705Goal and Background: Multiple Logit design is recognized as a pivotal tool for analyzing and selecting flight routes. Given the inherent complexities of this industry, including sudden fluctuations in passenger demand, technical and economic constraints, and stringent safety regulations and standards, the application of robust mathematical modeling for multiple logit design becomes imperative. models must possess the capability to manage uncertainties and analyze various scenarios to facilitate optimal route selection decisions. <br />Methodology: study focuses on designing an air multiple logit model using robust mathematical modeling for flight route selection. The proposed model effectively contributes to reducing costs and improving the efficiency of flight routes by incorporating various factors, including aircraft departure costs, flight crew wages, flight duration between two cities, and other parameters such as passenger capacity per route. The designed model was formulated and solved using GAMS software. The global optimal solutions for the three objective functions of the proposed model were calculated.<br />Results: Based on the obtained results, the first objective function is 182.8240 monetary units, the second is 540.6540 units, and the third is 200.369 monetary units. This model offers the necessary flexibility for optimal flight route management by accounting for uncertainties and market dynamics. The results indicate that implementing this robust model leads to an increase in aircraft departure costs with an increase in samples, while the total objective function value remains stable. Furthermore, flight crew wages remained constant with increasing samples, and this did not result in an increase in the total cost function valuehttps://jdaid.qom.ac.ir/article_4878_150b6639583170dccaa798174f2e6930.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Building Trust in Artificial Intelligence Deployment: Designing a Trustworthy Artificial Intelligence Framework Through a Mixed Method study212239487910.22091/jdaid.2026.16682.1087ENNazaninMohammadi HaroniyehDepartment of Management, Faculty of Management abd Accounting, Hazrat-e Masoumeh University, Qom, IranMonaJami PourDepartment of Management, Faculty of Management and Accounting, Hazrat-e Masoumeh UniversityJournal Article20260708Objective: With the rapid expansion of artificial intelligence (AI) applications in businesses, the issue of trust in this technology has become one of the fundamental challenges in its adoption and effective utilization. A lack of sufficient trust in AI systems can lead to reduced adoption, increased risk, and weakened competitive advantage in businesses. The present study aims to provide a comprehensive and localized framework for realizing trustworthy AI in businesses. <br />Method: To achieve the research objective, a mixed-methods approach was adopted. In the first phase, a comprehensive literature review, together with a qualitative approach, was conducted to identify the criteria and components of trustworthy AI. During the qualitative phase, data were collected through semi-structured interviews with experts and analyzed using thematic analysis. In the second (quantitative) phase, the Best-Worst Method (BWM) was employed to prioritize and assign weights to the five main criteria and 35 identified sub-criteria. The BWM calculations were performed using LINGO software to determine the final weights.<br />Results:The findings showed that achieving trustworthy AI in businesses requires simultaneous attention to technical and operational, human, ethical, legal, sustainability, and social impact dimensions, among which the technical and operational aspects are the most important. <br />Conclusions: These results indicate that despite the rapid growth of AI in businesses, there is still no practical, localized framework in Iran that considers all these aspects in an integrated manner. Based on this, the present study presented an integrated framework for realizing trustworthy AI that can be used as a practical guide for businesshttps://jdaid.qom.ac.ir/article_4879_7244e5f54e7754f4fcd0c1b0b74dbef9.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701Design of an Intelligent Predictive Framework for Financial Value Chain Dynamics and Stock Performance in Energy-Intensive Industries under Energy Price Fluctuations240255488010.22091/jdaid.2026.16733.1095ENKouroshKiyanfardDepartment of Industrial Management, AK.C., Islamic Azad University, Aliabad Katoul, IranMohammadTaghiKhosravi LarijaniDepartment of Management, AK.C., Islamic Azad University, Aliabad Katoul, IranAhmadMehrabianDepartment of Industrial Engineering, AK.C., Islamic Azad University, Aliabad Katoul, IranRaziehNaderkhaniDepartment of Mathematics and Statistics, AK.C., Islamic Azad University, Aliabad Katoul, IranJournal Article20260106Purpose: The industrial sector, particularly the cement industry, is highly energy-intensive, with Iran's consumption reaching 1.31 times the global average. Energy price liberalization and subsidy reforms have challenged resource management in this sector. This study aims to design an intelligent decision support system (DSS) to evaluate and predict the sustainable financial value chain of cement companies under energy price fluctuations.<br />Design/methodology/approach: This research integrates fundamental management principles with advanced technical components. Using time-series data mining and a sequential artificial neural network (ANN) architecture equipped with an advanced optimizer, the study models the hidden patterns of electricity and gas price fluctuations on corporate financial statements. KMO and Bartlett’s tests were employed to confirm sampling adequacy, while RSI and MACD indicators were adapted to capture price momentum and trends.<br />Findings: The findings reveal that changes in energy carrier prices directly affect industry returns and financial sustainability by altering price momentum and trend indicators. The proposed intelligent model demonstrated high accuracy in predicting the production chain's financial behavior, achieving a significantly reduced Mean Squared Error (MSE) and an exceptionally high coefficient of determination.<br />Originality/value: This study contributes to the literature by bridging technical data mining techniques with financial sustainability management. It offers a novel, highly accurate predictive framework specifically tailored for evaluating the financial resilience of heavy industries facing energy policy shifts.https://jdaid.qom.ac.ir/article_4880_d64fe8e8bb7a21febcc705302c49eabf.pdfUniversity of QomJournal of Data Analytics and Intelligent Decision-making3115-81612320260701An Integrated Fuzzy Expert-Based Decision Framework for Prioritizing Financial Risk Mitigation Strategies in Uncertain Business Environments256297488310.22091/jdaid.2026.17089.1115ENMajidAbdiAssistant Professor, Faculty of Management and Industrial Engineering, Malek Ashtar University of Technology, TehranJournal Article20260605Objective: This study introduces a fuzzy expert-based framework to prioritize financial risk mitigation strategies under uncertainty. By structuring qualitative expert inputs, the model bridges risk identification with strategic response, offering a transparent, analytical tool for complex financial environments. It enables managers to effectively address critical vulnerabilities and improve organizational resilience. Ultimately, this approach provides a robust, flexible methodology for intelligent financial management and serves as a strong foundation for future research.<br />Method: The proposed framework establishes a structured, fuzzy multi-criteria decision process to prioritize financial risk mitigation strategies. After identifying key risks and strategies through literature review and expert consultation, financial professionals provide linguistic assessments, which are modeled as fuzzy values to preserve uncertainty. The framework’s logic synthesizes these inputs by deriving risk importance and aggregating strategy performance. This integrated mechanism effectively handles conflicting expert judgments, generating a stable, interpretable ranking that serves as an intelligent prioritization tool for complex financial contexts.<br />Results: The framework successfully prioritizes financial risk mitigation strategies under uncertainty. Findings identify liquidity risk, inflation, and volatility as critical challenges, while intelligent forecasting, scenario planning, and dynamic cash-flow management emerge as primary mitigation responses. <br />Conclusions: This study proposes a fuzzy expert-based framework for prioritizing financial risk mitigation strategies under uncertainty. It connects risk identification with strategy evaluation in a transparent model that captures uncertainty and integrates differing expert views. The framework helps managers identify critical vulnerabilities, allocate resources more effectively, and improve financial resilience.https://jdaid.qom.ac.ir/article_4883_8b47cb0e933db644d503132fc3a1d239.pdf