abdollah hadi-vencheh

Sustainable economic growth is a key national priority, with bank loans serving as a critical driver by financing production units. However, rising Non-Performing Loans (NPLs) jeopardize economic stability, potentially triggering recessions. This study proposes an optimized loan allocation model for private banks, aiming to minimize NPLs while enhancing resource efficiency.

Using statistical techniques, including stepwise multiple regression, panel data analysis, and logistic regression, the study examines loan disbursement data, NPL ratios, and their determinants across three dimensions: bank-specific, firm-level, and macroeconomic factors.

The capital surplus-to-assets ratio, capital adequacy, financial soundness, and equity ratios significantly reduce NPLs and enhance allocation efficiency. At the firm level, industry sector, credit history, loan purpose, and banking relationship history all directly shape default risk, with industry type and credit history being the most critical factors in determining credit risk. Macroeconomic variables— including government debt, unemployment, economic growth, and the share of loans in investments—also systematically influence NPL trends and banks' resource allocation capacity.

Originality/Value: 

This research presents a comprehensive and actionable model for Iran's private banks, integrating multi-level indicators to optimize lending decisions and enhance credit risk management. The model equips bank managers with a strategic tool to improve operational efficiency and support economic stability.

This research presents an innovative framework that combines Hybrid Multi-Criteria Decision-Making (MCDM) approaches with Random Forest Regression to address interval-based fuzzy uncertainty in renewable energy project evaluation. Traditional Fuzzy TOPSIS methods often struggle with the inherent uncertainty and complexity of real-world data, which can lead to suboptimal decision-making. To enhance decision accuracy, we propose a hybrid solution that integrates Higher Interval TOPSIS with Random Forest Regression. This methodology effectively captures intricate interdependencies among project attributes—including cost, energy output, environmental impact, and social acceptance—within an interval-based fuzzy context. We applied our approach to a dataset of renewable energy projects and compared it against conventional Fuzzy TOPSIS methods. Results indicated significant improvements in predictive performance, achieving a Mean Absolute Error (MAE) of 0.045, a Mean Squared Error (MSE) of 0.0029, and an R² value of 0.95, highlighting the framework's ability to explain 95% of the variability in outcomes. This research underscores the promise of integrating AI-driven techniques within MCDM frameworks to enhance decision-making under uncertainty in the renewable energy sector.

In this paper, we propose a novel and straightforward nonlinear programming approach for aggregating individual composite indicators (CIs) into a group-level composite indicator (e.g., an aggregate CI for a group of entities). Drawing on performance measurement literature, our model is designed to be both simple and computationally efficient, requiring no specialized solvers for implementation. The proposed approach addresses the growing need for robust and interpretable methods to synthesize multidimensional data, particularly in contexts where policymakers and researchers aim to compare and benchmark the performance of groups or regions. To demonstrate the practical application of our method, we compute an aggregate Human Development Index (HDI) for the European Union (EU) region using HDI sub-indicators from individual EU member states. This case study highlights the model’s ability to integrate diverse dimensions of human development—such as health, education, and standard of living—into a single, coherent metric. By doing so, we provide a tool for evaluating the collective progress of the EU region while preserving the unique contributions of each member state. Our approach offers several advantages: (1) it is computationally accessible, making it suitable for a wide range of applications; (2) it allows for flexibility in weighting and aggregation, accommodating diverse policy priorities; and (3) it provides a transparent framework for constructing group-level CIs, enhancing their utility for decision-making and public communication.

Linguistic variables (LVs) provide a reliable expression of cognitive information. By inheriting the advantages of LVs, we can express uncertain and incomplete cognitive information in multiple attribute decision-making (MADM), and they do so better than existing methods.  In the decision-making process, we can consider decision experts’ (DEs’) bounded rationality, such as cognition toward loss caused by the DEs’ cognitive limitations during the decision process. Therefore, it is necessary to propose a novel cognitive decision approach to handle MADM problems in which the cognitive information is expressed by LVs. In this paper, we employ LVs to represent uncertain and hesitant cognitive information. Then, we propose a mathematical programming approach to solve the MADM problems where attributes or cognitive preferences are not independent.  Moreover, the validity and superiority of the presented approach are verified by dealing with a practical problem.

The objective of this manuscript is to introduce an innovative methodology for addressing multiple attribute group decision-making (MAGDM) problems utilizing interval-valued intuitionistic fuzzy sets (IVIFS). The proposed approach solves the problem using a mathematical programming methodology. In the present investigation, a group decision-making problem characterized by IVIF multiple attributes is conceptualized as a linear programming model and resolved expeditiously. The models that are being proposed have been reformulated into two analogous linear programming (LP) models through the application of a variable transformation and the concept of aggregation operators. The obtained LP models are solvable by common approaches. The principal benefit of the suggested methodology is its facilitation of decision-makers (DM) in identifying an alternative that exhibits optimal performance, and the decision-making process does not rely on DM knowledge. Application of the proposed method is represented in a decision-making problem, and the results are compared with similar methods, proving the compatibility of the proposed method with previous ones. The solid and understandable logic with computational easiness are the main advantages of the proposed method.

Considering that finding a suitable route in daylight hours and busy city with traffic restrictions is a big problem that not only causes non-optimal performance in distribution networks, in this regard, after modeling the problem in the form of vehicle routing development VRP) and considering the traffic and time window constraints and its NP-hard, using genetic metaheuristic algorithms (GA) and particle swarm optimization (PSO) to solve the problem and the optimal route and the number of vehicles required to send The product is specified. Customers' locations are first created using the clustering algorithm, location-based clusters, and sub-clusters according to the delivery time window, then a user interface receives the origin and destination provided by the user as input, this interface with Google Map Connection Receives directions between source and destination. Proposed routes are created using the proposed algorithms and using VANET network routing protocols, route events such as traffic are announced and, if necessary, the vehicle travels from the alternative route. The proposed method has better results than the optimal answers in terms of minimizing distance and number of vehicles.

This study aims to tackle the challenging facility location selection problem in Multiple Criteria Decision Making (MCDM) scenarios, explicitly focusing on type-1 fuzzy MCDM issues. The research introduces Interval Valued Fuzzy Numbers (IVFNs) to express ratings, addressing the difficulty in determining precise membership degrees for fuzzy sets.

The proposed IVF-COPRAS method, centered on uncertainty risk reduction, is employed to enhance decision-making reliability in IVF decision problems. This methodology is applied to a real-world case involving the selection of a location for municipal wet waste landfill pits in a major Iranian city. Comparative analyses with other methods are conducted to assess the proposed approach.

The study demonstrates the effectiveness of the IVF-COPRAS method in addressing facility location selection problems within MCDM. By utilizing IVFNs, the method successfully manages uncertainty, leading to more reliable decisions. Application to a practical scenario highlights the method's efficacy, and the comparative analysis provides insights into its performance relative to other methods.

Originality/Value: 

This research contributes a novel approach with the IVF-COPRAS method for handling facility location selection challenges in MCDM. The reliance on IVFNs offers a unique perspective on uncertainty in decision-making, enhancing decision reliability. The real-world application emphasizes the method's practical significance, providing a valuable contribution to MCDM research and offering a methodological tool for similar decision-making problems across diverse domains.

In today's industrial and competitive world, the optimal use of time, resources, and timely and correct response to market needs are crucial. On the other hand, the importance and necessity of optimizing production based on market demand will reveal the need to manage high-value-added products. Regarding this fact that the steel industry is one of the basic industries and has a great influence upon the other industries, its productivity can affect itself as well as the other industries. The scope of this study is Isfahan Steel Corporation which is considered as one of the greatest plants throughout the country manufacturing steel products such as iron beams. Thus in this study, the corresponding documents and proofs were examined to optimize the beam products of Isfahan Steel Corporation based upon the market demand and increased value, and considering the records, opinions of production and sale experts, customers, and suppliers, a questionnaire was prepared to estimate the types of beam and distributed among customers and suppliers and analyzed. Then, the added value of beams was calculated and finally, the number of beam products was calculated based on estimated market demand using the model ARIMA.

Preventive maintenance (PM) of machines has the critical role in a factory or enterprise. It decreases number of failures, increases reliability, as well as minimizes costs of production systems.  The managers’ duty of maintenance section is to prioritize machines and then, implement PM programs for them. Since machines have the different measures with respect to the maintenance costs, reliability, mean time between failures (MTBF), availability of spare parts, etc., the machines evaluation problem can be considered as a multiple criteria decision-making (MCDM) problem. Accordingly, the MCDM techniques can be applied to solve them. The aim of this paper is to extend the ELECTRE III (eLimination et choix traduisant la realite´– elimination and choice translation reality) method to interval type-2 fuzzy sets (IT2FSs) using curved (such as Gaussian) membership functions (MFs). The extended ELECTRE III methodology is then utilized to a maintenance group MCDM (GMCDM) matrix including the quantitative and qualitative criteria. In the proposed approach, the criteria weights, the assessment of alternatives with respect to criteria, and the thresholds are stated with Gaussian interval type-2 fuzzy sets (GIT2FSs). In order to show the effectiveness and applicability of the proposed approach, a case study and an illustrative example are exhibited using real decision-making problems. Due to the high correlation coefficients between our method and the others, as well as the results obtained by the proposed method, it can be taken into account as a valid and reliable approach to prioritize machines for PM.

Supplier selection is a process by which firms identify, evaluate, and contract with suppliers and plays a vital role in the management of a supply chain. Hence, the goal of the current paper is to propose a simple non-parametric model for the multiple criteria supplier selection problem. The suggested model provides non-zero weights for all criteria and allows the manager to get faster results by ranking the suppliers without solving the model n times.

Ranking DMUs based on individual preferences is an interesting and useful part of decision-making problems. Comparing the weighted sum of the selected number of rank votes, after determining the weights in a selected rank, can be regarded as a common approach to compute the total ranking of alternatives. In actual applications, making the weight of a certain rank zero means that we throw away the corresponding part of the obtained rank voting data. This paper proposes a new model to assess the non-zero weights for each position.

Cross-efficiency evaluation in Data envelopment analysis (DEA) has been accepted as a useful tool for performance evaluation and ranking of decision making units. In this paper using Undesirable Multiple Form (UMF) model with specific risk of α, a new stochastic model called Expected Ranking Criterion is introduced using statistical techniques for efficiency evaluation decision making units (DMU). Another issue in applying cross-efficiency DEA models is considering stochastic in input and output variables. Also, the non-uniqueness of optimal weights in this evaluation has reduced the usefulness of this powerful method. As a result, it is recommended that secondary goals be introduced in cross-efficiency evaluation. In this paper, the cross-efficiency model is modified to deal with stochastic data by applying chance-constrained approach.

Supplier selection plays an essential role in organizations due to the cost of raw material constitutes the main cost of the final product. Thus, we develop a new approach to solve the multiple criteria supplier selection problem. The proposed method considers the effects of weights in the final solution. An illustrative example is presented to show the capabilities of our approach.