Supply Chain Analytics最新文献

{"title":"A deep learning and policy optimization approach for supply chain order classification","authors":"Ramakrishna Garine ,&nbsp;Ripon K. Chakrabortty","doi":"10.1016/j.sca.2025.100166","DOIUrl":"10.1016/j.sca.2025.100166","url":null,"abstract":"<div><div>Timely delivery is a critical performance metric in supply chain management, yet achieving consistent on-time delivery has become increasingly challenging in the face of global uncertainties and complex logistics networks. Recent disruptions, such as pandemics, extreme weather events, and geopolitical conflicts, have exposed vulnerabilities in supply chains, resulting in frequent delivery delays. While traditional heuristics and simple statistical methods have proven inadequate to capture the myriad factors that contribute to delays in modern supply chains, Machine learning (ML) and Deep Learning (DL) approaches have emerged as powerful tools to improve the accuracy and reliability of delivery delay prediction. Consequently, this study presents a hybrid predictive framework that integrates DL models with Reinforcement Learning (RL) to improve binary classification of order status (on-time vs. late). We first benchmark several DL architectures, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Bi-LSTM, and Stacked LSTM, enhanced with regularization and extended training epochs, alongside a fine-tuned eXtreme Gradient Boost (XGBoost) model. These models are evaluated using accuracy, precision, recall, and the F1-score, with Bi-LSTM and Stacked LSTM achieving strong generalization performance. Building on this, we deploy a Proximal Policy Optimization (PPO) agent that incorporates deep learning outputs as part of its observation space. The RL agent uses a reward-based feedback loop to improve adaptability under dynamic conditions. Experimental results show that the hybrid DL-RL model achieves superior classification accuracy and an F1-score greater than 0.99, outperforming standalone methods. Although the PPO agent alone struggled with detecting minorities due to imbalance, integrating DL features mitigated this limitation. The findings support the use of hybrid architectures for real-time order status prediction and provide a scalable pathway for intelligent supply chain decision making. Future work will address class imbalance and enhance policy robustness through cost-sensitive and explainable RL strategies.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100166"},"PeriodicalIF":0.0,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An analytical framework for decision criteria validation in complex supply chains","authors":"Frank Michael Theunissen,&nbsp;Shafiq Alam,&nbsp;Aymen Sajjad","doi":"10.1016/j.sca.2025.100169","DOIUrl":"10.1016/j.sca.2025.100169","url":null,"abstract":"<div><div>Multi-Criteria Decision Making (MCDM) in supply chain management often applies rigorous methods for weighting and aggregation yet devotes little attention to the structural validity of the decision criteria that precede them. Even when organisations do not proceed to full MCDM model application, criteria are still elicited during problem structuring and used to justify initiative selection. This paper introduces a topological validation framework that addresses this asymmetry by representing criteria as a high-dimensional Decision Criteria Configuration (DCC). Using tools from Topological Data Analysis (TDA), we translate foundational MCDM axioms into measurable invariants: completeness through connectivity, non-redundancy through structural impact analysis, and logical consistency through cycle detection. Two industrial experiments demonstrate the framework’s utility. In a supply chain strategy-setting workshop, TDA diagnosed the criteria set underpinning initiative selection as a “conceptual monolith,” revealing significant redundancies and systemic feedback loops overlooked by conventional facilitation. In a subsequent inventory classification exercise, the audit resolved expert deadlock by reducing 32 proposed criteria to a minimal, non-redundant core of six operationally essential levers, providing an objective and defensible basis for moving forward. By transforming criteria sets into auditable decision architectures, this approach ensures that MCDM models and the initiatives they justify rest on a validated foundation before weighting or ranking alternatives. For managers, it functions as a pre-hoc “structural audit,” reducing redundancy, exposing hidden interdependencies, and directing resources toward criteria that genuinely drive strategic and operational outcomes.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100169"},"PeriodicalIF":0.0,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review of text mining analytics for supply chain risk management using online data","authors":"Georgios Gelastopoulos,&nbsp;Christos Keramydas","doi":"10.1016/j.sca.2025.100167","DOIUrl":"10.1016/j.sca.2025.100167","url":null,"abstract":"<div><div>Global supply chains are increasingly complex and vulnerable, requiring new approaches for detecting and managing risks. Text mining, a branch of natural language processing, can extract insights from unstructured online data such as news, reports, and social media. This paper presents a systematic review of 33 peer-reviewed studies on text mining in supply chain risk management (SCRM). The review addresses four research questions: (i) which types of online data are used and how their characteristics affect reliability and timeliness, (ii) which techniques are applied and with what trade-offs, (iii) how text mining contributes to risk identification, prediction, and mitigation, and (iv) what gaps and opportunities remain for future research. A bibliometric analysis is also conducted to highlight publication trends, contributors, and thematic clusters. The findings reveal that Twitter and news sources dominate, while methods range from sentiment analysis and topic modeling to advanced neural models such as BERT. Applications emphasize risk identification and visibility, with emerging work in predictive analytics and decision support. A conceptual framework is proposed linking unstructured data to risk management decisions. This review contributes to the literature by underscoring the value of real-time textual for improving visibility, agility, and resilience in complex supply chains.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100167"},"PeriodicalIF":0.0,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An analytical approach to blockchain-driven identity management in sustainable forest supply chains","authors":"Robertas Damaševičius ,&nbsp;Rytis Maskeliūnas","doi":"10.1016/j.sca.2025.100161","DOIUrl":"10.1016/j.sca.2025.100161","url":null,"abstract":"<div><div>This study explores the application of Self-Sovereign Digital Identity (SSDI) and blockchain technology in forest supply chain management to improve traceability, sustainability and regulatory compliance. It addresses how these technologies can overcome the limitations of traditional identity management and improve forestry operations’ transparency, efficiency, and environmental accountability. An Ethereum-based blockchain framework was used for this study, focusing on metrics such as transaction throughput and latency. Experimental tests were conducted to analyze the performance of SSDI in forest supply chains, focusing on real-time data management and secure identity control. A framework aligned with the Forest 4.0 initiative was proposed to evaluate the efficacy of SSDI. The results show that the integration of SSDI with blockchain significantly improves traceability and sustainability within forest supply chains, with high transaction rates and reduced latency. The decentralized system improves transparency and trust, promotes efficient identity management among stakeholders, and improves compliance with environmental regulations. Our study is among the first to apply SSDI in forestry, advancing digital transformation in this sector. Demonstrating SSDI’s capacity to streamline data handling and boost traceability, it offers practical recommendations for stakeholders seeking sustainable and digitally secure supply chain management practices.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100161"},"PeriodicalIF":0.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A textual analytics approach to sustainable supply chain dynamics in European maritime logistics","authors":"George R. Dimakis,&nbsp;George Tsironis,&nbsp;Konstantinos P. Tsagarakis,&nbsp;Yannis Marinakis","doi":"10.1016/j.sca.2025.100163","DOIUrl":"10.1016/j.sca.2025.100163","url":null,"abstract":"<div><div>This paper offers a comprehensive analysis of 871 European maritime firms, focusing on the spatial distribution of their headquarters, workforce demographics and digital footprint, as measured by LinkedIn follower metrics. To complement the quantitative data, Latent Dirichlet Allocation (LDA) was employed for text mining analysis on company LinkedIn descriptions, revealing emergent themes in innovation, customer-centric philosophy and global integration. The results point to a strong regional concentration of firms in Great Britain and the Netherlands, reflecting historical marine legacies and robust port infrastructures. Furthermore, the prevalence of small to medium-sized enterprises (SMEs) highlights the industry’s fragmented yet resilient structure, while digital presence remains uneven across firm sizes, with only a minority achieving substantial influence and visibility on social media. To summarize, these insights suggest that maritime logistics holds potential to drive systemic improvements in operational coordination, regional development, and global trade connectivity. Enhancing its integration could support more efficient supply chains, mitigate regional disparities, and bolster the industry’s global competitiveness.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100163"},"PeriodicalIF":0.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A deep reinforcement learning and fractional packing framework for routing and scheduling in healthcare waste supply chains","authors":"Norhan Khallaf ,&nbsp;Osama Abdel‑Raouf ,&nbsp;Mohiy Hadhoud ,&nbsp;Mohamed Dawam ,&nbsp;Ahmed Kafafy","doi":"10.1016/j.sca.2025.100164","DOIUrl":"10.1016/j.sca.2025.100164","url":null,"abstract":"<div><div>Artificial intelligence (AI) is increasingly utilized in healthcare logistics, including automated systems for collecting hazardous medical waste from hospitals under strict time and capacity constraints. This study compares three routing algorithms: (1) Google Maps Destination using an application programming interface (API), (2) hybrid clustering with Deep Q-Network (DQN), and (3) a hybrid method combining clustering, the fractional knapsack strategy, and DQN. These algorithms aim to optimize route planning and scheduling for medical waste collection vehicles operating under real-world constraints such as limited vehicle capacity and fixed service windows. The routing problem is modeled as both a capacitated vehicle routing problem (CVRP) and a CVRP with time windows (CVRPTW), capturing complexities. A multi-trip routing strategy is integrated into the promising algorithms to assess its impact on performance metrics, including capacity utilization, travel distance, total operational time, and number of trips. Experimental results indicate hybrid approach with clustering, fractional knapsack, and DQN outperforms others. It achieved capacity utilization rates of 96.47 percent for CVRP and 76.01 % for CVRPTW, requiring six vehicles, a 25 % reduction compared to the Google Maps API method, while matching the performance of clustering with DQN under time constraints. The CVRP model improved capacity utilization by 28.9 % over Google Maps API and 85.1 % over clustering with DQN. Although travel distance increased slightly (0.61 % in CVRP and 7.2 % in CVRPTW), total operational time was reduced by 7.6 and 4.6 %. The proposed model also minimized extra trips, requiring none for CVRP and only one for CVRPTW, compared to two additional trips needed by clustering with DQN in both scenarios. These findings highlight the hybrid approach as a robust, efficient solution for medical waste transportation under complex conditions.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100164"},"PeriodicalIF":0.0,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A scoping review and bibliometric analysis of sustainable and resilient supply chain network design","authors":"Rahmi Yuniarti ,&nbsp;Suparno ,&nbsp;Niniet Indah Arvitrida","doi":"10.1016/j.sca.2025.100162","DOIUrl":"10.1016/j.sca.2025.100162","url":null,"abstract":"<div><div>Designing sustainable and resilient supply chain networks (SRSCND) has become a strategic priority amid intensifying environmental pressures, market volatility, pandemic disruptions, and geopolitical uncertainties such as trade wars, resource nationalism, and regional conflicts. This study employs a hybrid bibliometric–scoping review (ScoRBA) combined with the PAGER framework to systematically map and synthesize 528 peer-reviewed articles published between 2015 and 2025. The analysis identifies five thematic clusters: (1) digitalization for sustainable decision-making, (2) energy and environmental priorities in low-carbon supply chains, (3) resilience and strategic planning under uncertainty, (4) value-oriented and data-driven reverse supply chains, and (5) heuristic optimization in green and closed-loop systems. Cross-cluster insights highlight that the most innovative solutions emerge at the intersections of these themes—for example, integrating digital decision-support systems with adaptive heuristic optimization for real-time network reconfiguration; coupling circular economy strategies with resilience planning to create low-carbon yet disruption-ready systems; and combining traceability infrastructures with value-recovery optimization in closed-loop networks. Although conceptual maturity is well established, operational maturity remains limited: most studies rely on theoretical modeling, simulation, or isolated case studies, with few sector-specific real-world applications. Social and behavioral dimensions, governance integration, and multi-sector disruption modeling remain underexplored. Future research should prioritize scaling pilot projects into multi-sector industrial implementations, embedding social, cultural, and behavioral factors into quantitative models, and developing adaptive real-time decision systems that integrate environmental, economic, and social objectives. Strengthening industry–academia collaboration, improving open-data access, and leveraging digital twin technologies will be critical to accelerate the transition from theoretical advances to scalable, practice-oriented solutions for building sustainable and resilient supply chains in an era of complex global risks.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100162"},"PeriodicalIF":0.0,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An integrated analytical framework for inventory and pricing of perishable products in multi-echelon supply chains","authors":"Jesús Isaac Vázquez-Serrano ,&nbsp;Leopoldo Eduardo Cárdenas-Barrón ,&nbsp;Julio C. Vicencio-Ortiz ,&nbsp;Neale R. Smith ,&nbsp;Rafael Ernesto Bourguet-Díaz ,&nbsp;Armando Céspedes-Mota ,&nbsp;Rodrigo E. Peimbert-García","doi":"10.1016/j.sca.2025.100157","DOIUrl":"10.1016/j.sca.2025.100157","url":null,"abstract":"<div><div>Inventory management and pricing strategies are fundamental to supply chain operations, particularly for wholesalers who serve as intermediaries between manufacturers and retailers at specific times. A wholesaler's profitability depends critically on two key operational decisions: effective inventory control to minimize costs and strategic price-setting for retail customers. This paper introduces an innovative hybrid model that combines optimization and discrete-event simulation to address these challenges, with a specific focus on perishable goods management and determining break-even pricing points. The proposed hybrid model is comprehensive in scope, accommodating multiple perishable products across various time periods and suppliers while accounting for the inherent uncertainties in wholesale operations. Its dual-component structure leverages optimization techniques for inventory cost minimization while employing simulation to address operational variability. The model provides detailed mathematical frameworks for calculating unit-level critical selling prices, both inclusive and exclusive of operational costs. To validate the model's effectiveness, the research presents a case study of a pharmaceutical wholesaler, drawing on data from the United Nations Office for Project Services. The hybrid model's performance was evaluated against two established empirical methodologies in the supply chain: the Lowest Acquisition Cost Approach and the Earliest Product Acquisition Approach. The results demonstrate significant improvements, with the hybrid model achieving a 20 % reduction in average total costs and an 18 % decrease in average critical selling price compared to traditional approaches.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100157"},"PeriodicalIF":0.0,"publicationDate":"2025-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145007564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comparative study of multi-algorithm optimization for inventory analytics in supply chains","authors":"Oussama Zabraoui,&nbsp;Yahya Hmamou ,&nbsp;Anas Chafi ,&nbsp;Salaheddine Kammouri Alami","doi":"10.1016/j.sca.2025.100154","DOIUrl":"10.1016/j.sca.2025.100154","url":null,"abstract":"<div><div>Effective management of inventory is essential for achieving high service levels, minimizing costs, and maintaining the overall resilience of retail supply chains—particularly in complex, real-world environments. Conventional strategies often prove inadequate because they rely on rigid assumptions or single-technique models that fail to accommodate practical challenges such as fluctuating demand, unpredictable lead times, and disruptions in supply.</div><div>To bridge this gap, our research undertakes a comprehensive comparison of multiple approaches — including Reinforcement Learning (RL), Genetic Algorithms (GA), Deep Learning (DL), Machine Learning (ML), and heuristic techniques — evaluated within a consistent and realistic testing framework based on the Walmart M5 dataset. This dataset offers a robust benchmark, containing multi-store, multi-item sales data that captures seasonal trends, event-driven demand variations, and price sensitivity. We introduce and evaluate an innovative hybrid methodology that combines a Genetic Algorithm with a Deep Q-Network (GA–DQN). The GA component conducts a broad, global search to optimize static inventory parameters such as reorder points and safety stock, while the DQN module learns adaptive, state-aware ordering strategies that can respond to dynamic, uncertain conditions. Our results show that this hybrid GA–DQN model achieves a significant improvement over a standalone DQN baseline—raising the service level from 61% to 94% and simultaneously lowering overall inventory costs. The framework we propose is modular and includes three key components: demand forecasting using Long Short-Term Memory (LSTM) networks to capture temporal sales patterns; GA-based optimization to fine-tune static policy parameters; and RL-driven adaptive control to support responsive, real-time ordering decisions. This integrated approach delivers a scalable, data-driven solution well-suited to the demands of modern retail supply chains, effectively addressing issues such as supplier unreliability, demand uncertainty, and the management of perishable goods.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"12 ","pages":"Article 100154"},"PeriodicalIF":0.0,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144902459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An evaluation of traceability dynamics in dairy supply chains through causal modeling in emerging economies","authors":"Shahab Bayatzadeh ,&nbsp;Hamidreza Talaie","doi":"10.1016/j.sca.2025.100156","DOIUrl":"10.1016/j.sca.2025.100156","url":null,"abstract":"<div><div>Traceability capability to track the history, location, and application of dairy products is crucial for ensuring food safety, quality, and transparency across supply chains. However, its development in emerging economies, particularly in Iran, remains limited due to infrastructural and technological challenges. This study addresses this gap by identifying and analyzing the key factors that influence traceability in Iran’s dairy sector, which plays a critical role in national nutrition and public health. Using a hybrid approach, the fuzzy Delphi method was first applied to refine a set of 19 factors extracted from the literature, validating 14 context-relevant elements based on expert consensus. Subsequently, the fuzzy DEMATEL method, designed to model causal relationships under uncertainty, was used to determine interdependencies among these factors. The results highlight food safety and quality, supply chain process management, data analysis and forecasting, and data integration as the most influential drivers of traceability. Meanwhile, competitive advantage, sourcing transparency, and environmental sustainability were found to be dependent outcomes. This research contributes a contextualized, expert-based framework tailored to the Iranian dairy industry and offers practical implications for improving transparency, reducing waste, and building consumer trust. The methodology and findings are transferable to other developing country contexts facing similar challenges.</div></div>","PeriodicalId":101186,"journal":{"name":"Supply Chain Analytics","volume":"11 ","pages":"Article 100156"},"PeriodicalIF":0.0,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144878409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}