https://aie.ut.ac.ir/ Advances in Industrial Engineering en daily 1 Mon, 01 Jun 2026 00:00:00 +0330 Mon, 01 Jun 2026 00:00:00 +0330 https://aie.ut.ac.ir/article_103583.html During the recent years, the companies in a wide range of industries have to design their activities in such a way to reduce the costs. A most popular way to reduce the costs in logistics is cross-docking. It is a strategy which is used to serve different purposes including the fast consolidation of received volume of commodities from suppliers, improving the responsiveness by shortening delivery lead time, reducing the inventory holding costs, eliminating spoilage costs of commodities, reducing transportation costs by employing full truck loading policy etc. The objective of this paper is to develop a mixed integer linear programming (MILP) model considering supplier selection and order allocation, perishability of commodities, group purchasing strategy and multi-agent scheduling into the well-known vehicle routing problem with cross-docking. Some small-sized test instances are applied to validate the new proposed model. A weighted-sum method is applied to solve small-sized instances. Then sensitivity analysis of the new proposed model is performed on the key parameters of the objective functions so that the supply decisions are evaluated while the parameters of the distribution costs are changed. Due to NP-hardness of the new proposed problem, two meta-heuristic algorithms including NSGA-II and MOPSO are applied to solve a wide range of instances. The obtained results by applying statistical hypothesis tests are compared through six different criteria. Also, an ordering technique that is called Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is applied to rank the meta-heuristic approaches. https://aie.ut.ac.ir/article_105667.html Selecting appropriate technologies in healthcare systems is a complex decision-making problem involving multiple, often conflicting criteria under uncertainty. In this paper, a novel fuzzy multi-criteria decision-making (MCDM) method, referred to as F-PSWCA, is proposed to enhance the criteria weighting process by incorporating historical performance trends within a fuzzy environment. The proposed method integrates fuzzy regression parameters, including slope, intercept, and coefficient of determination (R²), to capture both the magnitude and stability of criteria over time. Unlike conventional fuzzy MCDM approaches that rely solely on static expert judgments, F-PSWCA enables dynamic assessment of criteria importance while preserving uncertainty representation. The applicability of the proposed method is demonstrated through a real-world case study on the selection of dialysis water purification technologies, where multiple technical, economic, and operational criteria are considered. Comparative analysis with Fuzzy SAW, Fuzzy TOPSIS, and Fuzzy SECA is conducted to evaluate the robustness and consistency of the results. The findings indicate that while ranking similarities may occur across methods, F-PSWCA provides additional interpretive insights by distinguishing between temporally stable and unstable criteria. The results confirm the effectiveness of the proposed approach as a decision-support tool for technology selection in complex and evolving healthcare environments. https://aie.ut.ac.ir/article_103942.html This study seeks to address gaps in previous research by introducing a comprehensive data-driven distribution network design model. The process begins with an in-depth analysis of customer demand, utilizing unsupervised learning algorithms to gain valuable insights into consumer behavior. This analysis identifies demand levels across different geographical regions and reveals temporal demand patterns. The resulting insights serve as inputs to the distribution network design model. To facilitate effective data classification and analysis, The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is employed. enabling accurate estimation of customer demand based on innovative parameters. Based on the clustering results, a mixed-integer linear programming model is developed that incorporates facility-location, capacity-planning, product-flow, and fleet-composition decisions. Importantly, during this modeling process, emphasis will be placed not only on optimizing the number, location, and capacity of facilities but also on refining fleet types and their compositions to enhance overall efficiency. The proposed model is solved using CPLEX in GAMS and evaluated through a set of numerical test instances. The results demonstrate that the proposed data-driven model achieves an average profit improvement of 10-15% compared to traditional non-clustered approaches. The model also yields savings in transportation and fleet-related costs. Moreover, its integrated structure enables sensitivity analyses of key parameters and provides useful managerial insights. demonstrating the synergy between data-driven clustering and mathematical optimization for distribution network design. https://aie.ut.ac.ir/article_105668.html Typically, humanitarian agencies (HAs) tend to preposition relief supplies before potential disasters to increase resource availability in the post-disaster phase. However, prepositioning imposes high costs on relief chains. On the other hand, although postponing relief supply procurement to the post-disaster phase may avoid prepositioning costs, it comes with significant supply risks. This study utilizes buyback, option, and quantity discount contracts which enables HAs to adopt a hybrid strategy for providing relief supplies, ensuring maximum post-disaster supply availability. In this paper, a robust bi-objective two-stage stochastic programming model is proposed for supplier selection and the procurement of critical and non-critical relief supplies. The model also emphasizes suppliers' profits to incentivize their participation in the relief chain. In this study, a combination of robust stochastic programming and robust convex programming is employed to manage uncertainty. The model is applied to a case study, and multiple sensitivity analyses are performed. The results demonstrate that, compared with the situation in which hybrid contracts are not utilized, the proposed model reduces the HA’s cost by 38.4% and expected shortages by 22.6%, while increasing suppliers’ profits by 41.8%. https://aie.ut.ac.ir/article_106090.html This article proposes a mathematical model to examine service pricing in ride-hailing companies and the relationship between ride-hailing and insurance companies, in which insurance is viewed as a competitive advantage. The ride-hailing company insures its drivers, customers, and cars with the insurance company, ensuring that the insurance company will compensate the affected parties in the event of an accident during travel. This study addresses the importance of hygiene in ride-hailing cars, given the impact of epidemics such as COVID-19 on transportation service prices. The interaction between the ride-hailing and insurance companies is modeled using a Stackelberg game in three scenarios: decentralized, centralized, and a coordination game under revenue-sharing contracts. The ride-hailing company decides on insurance and hygiene-level, while the insurance company determines the base entrance premium to interact with the ride-hailing company. This model allows ride-hailing companies to optimize their profitability and decision-making, and provide safe and reliable services to their customers. The study is validated using information from Lyft. The result shows that the revenue-sharing contract between the ride-hailing company and the insurance company increases the profits of both companies. This article provides a framework for ride-hailing companies to gain a competitive advantage by coordinating with insurance companies and ensuring the safety and hygiene of services. https://aie.ut.ac.ir/article_106335.html Performance evaluation in complex service systems is a central concern in industrial engineering, particularly in regulated environments such as healthcare. This study develops a data-driven industrial engineering framework for hospital performance evaluation structured through the Balanced Scorecard (BSC) and supported by exploratory predictive modelling. A structured evidence-based screening process was conducted to identify operationally measurable performance indicators, which were subsequently organised within the four BSC perspectives and aggregated into composite performance dimensions using standardised equal-weight scoring. Quarterly organisational data (2018–2024) from a tertiary hospital were used to examine structural relationships among indicators through a multilayer perceptron neural network. The predictive component is intended as exploratory validation rather than universal forecasting. Results indicate strong alignment between predicted and observed composite performance scores (R = 0.84), suggesting that the selected indicators collectively explain substantial variation within the studied organisational context. Importance and sensitivity analyses further identify cost-efficiency and operational-process variables as influential drivers, while safety, workforce, and patient-related indicators demonstrate meaningful associations. The proposed framework enhances transparency in indicator selection, clarifies multidimensional performance structuring, and provides analytically informed decision support for complex service systems. https://aie.ut.ac.ir/article_106336.html This study presents a comprehensive and robust framework for evaluating and ranking the energy and environmental efficiency of European Union countries under conditions of data uncertainty. A non-radial Slack-Based Measure model is proposed, incorporating undesirable outputs and utilizing a unified efficiency frontier. To rank countries based on interval efficiency results, a preference-based algorithm is employed that utilizes pairwise comparisons and incorporates the decision-maker's behavioral preferences through an optimism parameter. This approach enables flexible, behavior-sensitive rankings. Furthermore, the study demonstrates that several commonly used possibility-degree-based ranking formulas yield equivalent results within the conventional interval comparison framework, regardless of whether intervals overlap or are completely disjoint. Moreover, the study illustrates that the proposed model offers significant advantages over these formulas in terms of discriminatory power, consistency, and robustness. To validate the proposed model, interval data covering the energy and environmental efficiency of 27 EU countries during 2021–2022 are analyzed. The results indicate that the model achieves high discriminatory power, consistent rankings, and robustness against data fluctuations. Empirical findings reveal that countries such as Luxembourg, the Netherlands, and Ireland exhibit superior energy and environmental performance, whereas countries like Romania and Poland rank significantly lower. Overall, the proposed framework serves as an effective tool for decision analysis and policymaking in the field of energy and environmental management.  https://aie.ut.ac.ir/article_106091.html Recording and classifying commerce data electronically within the new age of information and development has many benefits for sellers and consumers in the online supply chain. We can predict customer purchase behavior patterns by analyzing this classified data. In recent years, digital stores have received additional attention due to their advantages. On the other hand, the performance of these stores is directly associated with the performance of their suppliers. Hence, supply chain management is essential during this sales system. In this investigation, the classification methods (WFRM), demand prediction analysis (Binary Logistic Regression), data classification (Discriminant Analysis), time series analysis (Trend Analysis), and mathematical modeling have been used to select suppliers to order led to the development of a management strategy to prevent shortages and reduce the average inventory and costs of sending and producing for suppliers. It ultimately covers a 29% error in sales data for assigning suppliers to orders. https://aie.ut.ac.ir/article_104369.html Rapid urban growth, driven by urbanization and agglomeration of goods and services in metropolitan cities, jeopardizes land conversion for agriculture and challenges traditional models of urban development. The traditional urban growth models emit carbon, destroy the environment, and create food deserts, thereby compromising food security and health for urban residents. Sustainable urban agriculture can be an alternative solution since it provides food security and access to fresh and affordable food. The present research employs a two-stage model: regional ranking with an integrated ANP-TOPSIS method, and best allocation of resources, such as location selection, cultivation technology, and gender-balanced human resource deployment, with the mixed integer programming methodology. Implemented in Iran, Kermanshah, the proposed model recommends vertical hydroponic production of cauliflower and tomatoes to reduce water and land use. It also promotes gender balance in working opportunities, thus lessening the disparity among men and women in job activities. Economic evaluation using the net present value method reaffirms the economic viability of urban farming, with a mention of the effect of land price and return on investment. Analysis of the ANP-TOPSIS model establishes that increasing the level of sustainability raises the farm's sustainability, albeit non-linearly in all the dimensions and sub-criteria. This approach supports observations regarding effective urban agriculture practices towards sustainable urban development. https://aie.ut.ac.ir/article_104569.html Despite growing interest in digital transformation and modular services, many logistics firms—especially in Iran—lack a cohesive framework that integrates modular architecture with operational and technological needs amid rising complexity and customer demands. This study aims to develop a conceptual framework for a modular logistics service platform that enhances flexibility, innovation, and collaboration across supply chain actors. The research identifies core service modules and examines how modularity can support the design of efficient, adaptive service offerings. Using a qualitative case study approach, the study investigates one of Iran’s leading courier companies. Data were collected through semi-structured interviews with ten senior managers, direct observations, and analysis of internal documents. Thematic content analysis revealed key service modules, processes, and a three-layer modular architecture consisting of service, process, and activity layers. These are structured around the First Mile, Mid Mile, and Last Mile segments, incorporating nodes, links, and carriers as core elements. The platform supports modular processes such as routing and packaging, enables outsourcing at multiple levels, and integrates Artificial Intelligence (AI), and Internet of Things (IoT) technologies to optimize performance. The framework addresses major gaps in existing literature, including role definition, modular governance, smart technology integration, and service scalability. This research offers a novel, multi-level modular logistics framework validated in a real-world context, providing a practical blueprint for logistics firms seeking to transition to flexible modular platforms that enhance efficiency and collaboration. https://aie.ut.ac.ir/article_104918.html The optimal use of natural resources and energy has become an important issue and has been given attention worldwide due to the increase in population and environmental pollution. In recent years, Eco-industrial parks (EIPs) have gained popularity as a way to make better use of natural resources. In these parks, companies try to cooperate with each other by exchanging materials and energy and pay more attention to environmental issues. Unlike previous models proposed in this field, which were based on existing EIPs and were presented for improvement, we present a new model for creating these parks. In this study, we propose a mixed integer programming (MIP) model, considering real and feasible exchanges, which, taking into account sustainability conditions (economic, social, and environmental), selects industries to establish the park so that the value of exchanges is maximized and the costs of infrastructure construction are minimized. The results show that the selected set of industries leads to economic benefits, where the total value of exchanges exceeds the costs of infrastructure, thus supporting profitability and sustainability. https://aie.ut.ac.ir/article_105617.html From an economic perspective, efficiency refers to the optimal use of resources to produce the maximum possible output, highlighting its critical role in management systems. Managers generally strive to utilize resources effectively to maximize outputs and satisfy all stakeholders. This is particularly important in the Iranian banking industry, where high costs of fund maintenance make efficiency a central concern. In this study, considering the operational structure of Iranian banks, a five-sector model is proposed to illustrate the flow of activities within banks. Based on this structure, a mathematical model using Data Envelopment Analysis (DEA) is developed to evaluate the efficiency of the five sectors. A fuzzy-based approach is then introduced to solve the model. Application of the proposed model to 210 branches of a major Iranian bank indicates that, while the Services and Mobilization of Resources sectors demonstrate relatively high efficiency, the Management sector suffers from significant inefficiency. These findings highlight the need for focused managerial interventions to enhance operational performance in bank branches. https://aie.ut.ac.ir/article_106904.html Cancer is a complex and dynamic disease capable of rapidly spreading throughout the body and impairing the immune system. Although immune cells can directly attack tumor cells or activate other components of the immune response, their activity alone is insufficient to achieve complete tumor eradication. Consequently, from a healthcare systems engineering perspective, optimized treatment planning and scheduling are essential to maximize clinical efficacy and resource efficiency. This study employed MATLAB Simulink simulations to model the system dynamics and investigate the effects of chemotherapy and immunotherapy, both individually and in combination, on cancer and immune cells. Cellular population dynamics were first analyzed in the absence of treatment, followed by separate and combined evaluations of each therapy, with particular attention to cellular interactions and cancer cell drug resistance. This approach enables a systematic evaluation of treatment as a multi-objective optimization problem, balancing tumor clearance and immune preservation. The findings revealed that in patients with small initial tumor size and robust immune function, the immune system alone is capable of eliminating cancer cells without therapeutic intervention. However, in patients with large initial tumor size, the combined application of chemotherapy and immunotherapy was predicted to be effective, achieving complete tumor clearance within 38 days while preserving overall health. Collectively, the simulation results suggest that combined chemo-immunotherapy, especially when the sequence of administration is carefully considered, is predicted to be the most effective strategy among the tested protocols and may serve as a model‑based decision support tool for personalized healthcare management, pending prospective clinical validation. https://aie.ut.ac.ir/article_106905.html Container drayage is a key component of maritime intermodal logistics, yet it is associated with high operational costs and considerable greenhouse gas emissions due to the extensive use of diesel trucks. This study develops an optimization framework for container drayage operations using hybrid electric trucks. We introduce the Hybrid Electric Container Drayage Problem (HECDP), which incorporates multiple propulsion modes, pure electric, internal combustion, charging, and boost, within a unified routing and scheduling model. A mixed‑integer linear programming formulation is proposed to jointly determine vehicle routes, pickup–delivery operations, propulsion‑mode selection, and battery state‑of‑charge dynamics while satisfying container loading and operational constraints. The model is first validated through a small numerical example and then applied to a real-world case study based on the Inland Container Terminal (ICT) Tilburg in the Netherlands. Computational experiments compare three fleet configurations: a conventional diesel fleet, a fully hybrid fleet, and a mixed fleet scenario. Results show that a fully hybrid fleet can reduce operational costs by about 30% and CO₂ emissions by nearly 60% compared with the diesel benchmark. Furthermore, even a partial transition to hybrid trucks (e.g., 30% fleet penetration) captures a substantial portion of these benefits. Sensitivity analyses examining battery capacity, travel speed, and terminal location indicate that operational conditions significantly influence system performance. Overall, the findings suggest that integrating propulsion characteristics of hybrid trucks into routing decisions can simultaneously improve operational efficiency and environmental performance in container drayage systems. https://aie.ut.ac.ir/article_106906.html This study investigates the role of quality culture in enhancing continuous improvement and decision-making within Iran's agricultural sector, using Behrooyesh Pars Company as a case study. Focusing on key pillars—management commitment, customer orientation, and fact-based thinking—it reveals a strong empirical link between quality culture and operational performance. Notably, fact-based decision-making shows a 60% positive correlation with continuous improvement, underscoring their mutual reinforcement when embedded in a robust cultural framework. The research also demonstrates that management support and customer focus significantly influence decision-making quality and organizational adaptability. Unlike prior studies that primarily conceptualize quality culture, this work offers empirical insights into how these cultural elements directly impact operational outcomes. Ultimately, the study proposes a practical framework tailored for agricultural input manufacturers, connecting quality culture values to measurable outcomes such as improved decision-making, continuous improvement, and enhanced organizational performance. This research advances the understanding of quality culture’s practical benefits in agricultural operations. https://aie.ut.ac.ir/article_106907.html Organ transplantation logistics demand rapid, reliable, and medically viable delivery solutions, especially when relying on commercial flight networks. The viability of an organ is severely constrained by its cold ischemia time—the duration an organ can remain functional outside the human body under refrigeration. This study formulates deterministic and robust mathematical models to optimize organ transport routing under uncertainty. The transportation network is modeled as a directed graph, where airports serve as nodes and scheduled flights as arcs, integrating constraints such as flight eligibility, connection handling time, and air traffic control prioritization. Building on a deterministic Resource-Constrained Shortest Path (RCSP) model, a model using Mulvey et al.'s robust optimization approach is extended. This allows the model to account for various real-world disruptions, including departure delays and scheduling inconsistencies, using multiple predefined scenarios. Furthermore, a penalty-based objective function is used to minimize arrival time and deviation across scenarios, ensuring stable and feasible routing decisions. The model is implemented via CPLEX, with experiments conducted on synthetic data for multiple organ types. Results demonstrate that the robust model, while more computationally intensive, yields significantly more resilient and feasible solutions across delay scenarios. Sensitivity analyses further highlight the critical roles of connection time and flight segment penalties. Overall, the proposed model offers a practical, scalable decision-support tool for transplant networks facing operational uncertainty. https://aie.ut.ac.ir/article_106908.html Persistent imbalances between electricity supply and demand remain a significant challenge in many developing economies, where energy-intensive industries exert considerable pressure on power systems. Demand response (DR) programs, particularly Time-of-Use (TOU) pricing, are widely implemented to encourage load shifting and mitigate peak demand. However, in heavy industries such as steelmaking, the effectiveness of TOU tariffs is strongly influenced by operational feasibility, production continuity requirements, and the imprecision of production-driven energy needs. This study develops an optimization-based framework for industrial demand response scheduling using operational and tariff data from the Iranian Alloy Steel Company. The deterministic formulation is first developed as a baseline model to minimize electricity costs while satisfying operational and production constraints. The proposed approach then extends this formulation within a credibility-based possibilistic programming framework, where production-driven uncertainty in daily energy requirements is represented using triangular fuzzy numbers. The results indicate that implementing the possibilistic formulation improves operational reliability by increasing the off-peak load share from 35% to 42% under uncertain production conditions, incurring only a marginal cost increase of 4.85% compared to the baseline deterministic model. Furthermore, the credibility-based schedule reduces the peak load exposure by 5.7% (from 140 MWh to 132 MWh), avoiding excessive peak-hour concentration. Overall, the findings quantify a clear trade-off, revealing that a 4.85% investment in an operational safety margin yields a more balanced and secure schedule, significantly reducing the risk of severe cost escalations in steelmaking demand response. https://aie.ut.ac.ir/article_107001.html The main objective of this research is to design the integrated dimensions of the marketing system. Marketing management faces ambiguity in the boundaries of the discipline and a weakening of its strategic position within organizations, and the lack of an integrated framework has hindered effective planning and the formulation of competitive strategies. This research was conducted with a qualitative approach using a systematic literature review. Due to the complexity of marketing management concepts, initial categories were first identified through a systematic review. Then, using the fuzzy Delphi technique in two stages, the final indicators and components were extracted and refined. In the first stage of the fuzzy Delphi, 18 initial categories were reduced to 11 groups, and in the second stage, with expert consensus, 8 final dimensions were confirmed. The research findings led to the design of a comprehensive model that encompasses the essential components for effective marketing management. These components include: Process and Innovation, Structure and Culture, Assets and People, Technology, Environment and Business Analysis, Stakeholder Interactions, Governance and Strategy, and Value and Performance. In conclusion, by designing the integrated dimensions of the marketing system and applying the fuzzy Delphi method to refine the components, this research has taken an effective step toward systematizing marketing activities. The proposed framework, by creating a common language, transparent processes, clear roles, and comparable criteria, facilitates professional learning and helps organizations, through a correct understanding of the central role of marketing, to formulate more coherent planning and strategy. https://aie.ut.ac.ir/article_107014.html Building Information Modelling (BIM) has emerged as a key enabler of digital transformation in the Architecture, Engineering, Construction, and Operation (AECO) industry. However, the transition toward higher levels of BIM maturity, particularly Level 3, remains a significant challenge due to the complexity of integrating technological, organizational, and regulatory dimensions. The study proposes an integrated framework for evaluating BIM Level 3 implementation based on three core dimensions—technology, process, and policy—and examines its impact on key stakeholders from a Project Management Body of Knowledge (PMBOK) perspective. The research adopts a mixed-method approach, combining qualitative analysis and quantitative evaluation through structured questionnaires. A case study was conducted within the Tehran Construction Engineering Organization to validate the proposed framework. The results indicate that while technological readiness is relatively advanced, significant gaps persist in process integration and policy support. Furthermore, the findings highlight the critical role of stakeholder coordination and information flow in achieving effective BIM implementation. The proposed framework provides a practical decision-support tool for project managers and policymakers, enabling a more holistic understanding of BIM maturity and its implications. https://aie.ut.ac.ir/article_107792.html The global transition to sustainability places firms at a strategic crossroads, compelling them to balance profitability with Environmental, Social, and Governance (ESG) performance. This study confronts a critical challenge within this transition: the risk of internal cannibalization, where a firm's new green products erode the market for its conventional offerings. We introduce a novel two-phase, game-theoretic framework to analyze the dynamic interactions between duopolistic firms-an incumbent and a challenger-and a government regulator. Our analysis reveals that cannibalization is not an inevitable market force but an endogenous strategic variable that can be managed. We demonstrate that while calibrated policy support is crucial for catalyzing eco-innovation, overly aggressive interventions can paradoxically trigger a "crowding-out" effect, stifling private R&D incentives. The results show that the optimal path to aligning profitability with sustainability lies in a strategic dual-channel portfolio, where adaptive pricing transforms cannibalization from a systemic risk into a market capture mechanism. This research offers quantifiable policy thresholds to guide regulators and provides actionable strategies for corporate leaders to secure a competitive advantage in the green economy. https://aie.ut.ac.ir/article_107793.html he Network-Interdicted Vehicle Routing Problem (NIVRP) extends classical VRP settings by incorporating network interdiction, where selected arcs or nodes may be disrupted or rendered inaccessible. Consequently, routing decisions must account for potential interdictions. This setting induces a strategic conflict between Logistics Service Providers (LSPs) and an adversarial interdictor seeking to disrupt routes by targeting specific arcs. Typically, the interdictor lacks complete prior information regarding depot locations and customer demands, and acquiring such information is costly. Through repeated interactions, however, the interdictor can observe selected routes and seized cargo on interdicted arcs, gradually refining demand estimates and reducing uncertainty. Over time, this process enables increasingly accurate predictions of illicit cargo flows. The interaction is modeled as a repeated game, in which the interdictor engages in an online learning process to adapt to dynamically changing checkpoint configurations. This study analyzes this conflict within a repeated game framework using an online learning approach. A bi-level metaheuristic algorithm, leveraging semi-bandit feedback, is proposed. Computational experiments on randomly generated instances demonstrate that the proposed learning-based method significantly outperforms benchmark approaches and exhibits satisfactory convergence behavior. https://aie.ut.ac.ir/article_107794.html Banks and other organizations in the financial sector, particularly payment service providers (PSPs), require accurate forecasts of transaction counts to optimize procurement planning for point-of-sale (POS) devices under budget constraints. To address this need, this study proposes a two-level hierarchical forecasting framework for POS transaction counts to support tactical decision-making. At the first level, a time-weighted Long Short-Term Memory (LSTM) model is developed to forecast monthly transaction counts over a one-year horizon. A set of candidate exogenous variables is initially identified and subsequently refined using Goal Programming Best-Worst Method (GP-BWM) to improve model performance. At the second level, a Random Forest (RF) model is employed to estimate the relative weights of daily transactions within each month based on intrinsic daily characteristics. A forward feature selection approach is further applied to identify the most influential daily features affecting transaction distribution. The proposed framework is validated through a real-world case study of a PSP in Iran. The results indicate that the time-weighted LSTM achieves high predictive accuracy with R² values of 0.90 and 0.89 on validation and test datasets for monthly forecasting, respectively. Additionally, the daily feature selection process demonstrates strong explanatory power with an R² of 0.9625. These findings confirm the effectiveness and practical applicability of the proposed framework in supporting data-driven planning. https://aie.ut.ac.ir/article_108138.html Among the responsibilities of senior managers in each project-based organization, project portfolio selection (PPS) is critical. Practical optimization and decision-making models play a vital role in facilitating managerial decision-making. In this respect, the current study develops a new data-driven multi-criteria decision-making (MCDM) model for group decision-making in the PPS within the framework of project portfolio management. Mining-specific criteria will help enhance project evaluation and selection, allowing the project manager to make more accurate and efficient decisions. By incorporating interval-valued fuzzy sets (IVFSs) into the decision-making process, this paper addresses uncertainties and ambiguities in experts' judgments, making evaluations more realistic and flexible. A new interval-valued fuzzy weighting method, IVF-LMAW, is developed to determine the relative importance of the criteria, and a new IVF method, IVF-AURA, is presented to prioritize projects, offering higher accuracy and stability than conventional decision approaches. To validate the accuracy and robustness of the model, the criteria weights have been subjected to a sensitivity analysis, and the resulting rankings have been examined. To highlight the practical applicability of the model, a real case study in the mining industry has been performed on a set of projects. The results show that the proposed approach can efficiently, transparently, and reliably support the PPS process by simultaneously considering both financial and non-financial dimensions.