Computers & Industrial Engineering最新文献

{"title":"An integrated terminal allocation and berth allocation problem in multi-user and dedicated terminals considering container flow movement","authors":"Liming Guo ,&nbsp;Zhongquan Jia ,&nbsp;Jianfeng Zheng ,&nbsp;Jian Du ,&nbsp;Kjetil Fagerholt ,&nbsp;Baoli Liu","doi":"10.1016/j.cie.2025.111551","DOIUrl":"10.1016/j.cie.2025.111551","url":null,"abstract":"<div><div>Due to the emergence of dedicated terminals rented by liner shipping companies, terminals at a transshipment hub port can be divided into the multi-terminal with general berths and the dedicated terminal with dedicated berths, which can be distinguished in terms of the service recipient, handling efficiency, terminal capacity and container transportation convenience. This paper proposes an integrated terminal allocation and berth allocation problem (TABAP), where the terminal allocation problem (TAP) is to allocate the terminal (i.e., dedicated terminal or multi-user terminal) for all incoming vessels appropriately, and the extended BAP determines berthing times and berthing positions for vessels under the limitations of different service rules of dedicated and multi-user terminals. From maritime practice considerations, we further introduce the impact of import, export and transshipment container flow movements within and between terminals on operating costs and berthing times in the TABAP. Then, we present a new programming model for the TABAP and propose a column generation (CG) algorithm to obtain high-quality solutions for large-scale instances within an acceptable time. Numerical experiments are provided to validate the efficiency of the CG algorithm and the effectiveness of the proposed model.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111551"},"PeriodicalIF":6.5,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ECI-SCBC: An Efficient Collaborative Interaction method for Supply Chain Based on Blockchain Consensus Mechanism","authors":"Jinli Wang ,&nbsp;Hang Zhao ,&nbsp;Mengzhuang Liu ,&nbsp;Zehui Yuan ,&nbsp;Libo Feng ,&nbsp;Yimin Yu ,&nbsp;Hui Li ,&nbsp;Shaowen Yao","doi":"10.1016/j.cie.2025.111565","DOIUrl":"10.1016/j.cie.2025.111565","url":null,"abstract":"<div><div>In the field of intelligent manufacturing supply chains, there are issues such as difficult node collaboration, non- transparent information transmission, and high management costs. To address these challenges, this paper proposes an Efficient Collaborative Interaction Mechanism for Supply Chain Based on Blockchain Consensus Algorithm (ECI-SCBC). First, based on the Delegated Proof of Stake (DPoS) consensus algorithm, a vote weight constraint model is established by introducing counting weight constraints, enabling multiple nodes to participate collaboratively in voting events. Second, a DPoS-based incentive model is developed, introducing a dual-reward mechanism that motivates nodes through rewards for both supply chain and blockchain-related contributions. Third, the reputation model, voting mechanism, and incentive mechanism are integrated and applied in DPoS, forming RVI- DPoS. This mechanism enables efficient collaboration among nodes and trustworthy transmission of information. Experimental results show that the node participation rate reaches 87%, the error rate is reduced to 9%, and the transaction processing capacity (TPS) achieves 2,637 units per second. These findings indicate that employing advanced blockchain- based collaborative interaction methods and optimizing resource utilization can significantly enhance system performance, efficiency, and security. Furthermore, the proposed model demonstrates the potential to achieve efficient collaboration and information transparency in complex and dynamic real world supply chain environments.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111565"},"PeriodicalIF":6.5,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prescriptive strategy selection in station-based car sharing: Profit optimization using reinforcement learning","authors":"Jonas Rikowski ,&nbsp;Felix Weidinger ,&nbsp;Melanie Reuter-Oppermann","doi":"10.1016/j.cie.2025.111561","DOIUrl":"10.1016/j.cie.2025.111561","url":null,"abstract":"<div><div>Many car sharing operators struggle to operate at profit. In previous work, it has been shown that trip selection is an important lever to make car sharing systems more profitable and efficient by enabling them to automatically decide whether to accept or reject customer requests. By actively selecting customer requests to be served, car sharing providers are able to implement operative strategies to increase profit, namely increasing collected base fares, increasing collected time-/distant-dependent fares, or reducing operational costs such as relocation costs. However, trip selection has mostly been investigated assuming knowledge of future demands for a fixed horizon and can thus not be unrestrictedly applied in real-time. This paper suggests a prescriptive algorithm, namely a deep reinforcement learning approach (RLA), to solve the trip selection problem solely based on real-time information. Being a machine learning approach, RLA learns which of the above strategies has the highest potential, without the need to specify the booking regime in advance. Based on a simulation of a real car sharing system, we can show that the novel approach can resemble the general structure of optimal offline solutions that were obtained given global information in large parts. It significantly outperforms solutions provided by the state-of-the-art approach widely applied in practice and shows robustness by maintaining the lead even under changing conditions. Narrowly analyzing the solutions found by the novel approach, we apply it to answer complex managerial questions like determining promising relocation rates.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111561"},"PeriodicalIF":6.5,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-agent mutual trust evaluation and regulation mechanism for human-robot-XR collaborative assembly system","authors":"Yali Liu ,&nbsp;Jiacheng Xie ,&nbsp;Xiaojun Qiao ,&nbsp;Rui Du ,&nbsp;Xuewen Wang","doi":"10.1016/j.cie.2025.111558","DOIUrl":"10.1016/j.cie.2025.111558","url":null,"abstract":"<div><div>Under the context of Industry 5.0, the rapid advancement of frontier technologies such as Extended Reality (XR) and Artificial Intelligence is reshaping human-robot relationships. These relationships are shifting from traditional master–slave control structures toward more equal and collaborative partnerships. In complex and dynamic assembly environments, with increasingly intricate information flows and diverse interaction modalities, trust between human and robot has emerged as a key factor supporting efficient collaboration and stable system operation. To address this, the present study focuses on the structural characteristics of a human-robot-XR integrated assembly system and categorizes decision-making entities with autonomous perception and interaction capabilities into three types of intelligent agents: Agent_H (Human), Agent_R (Robot), and Agent_C (Computing devices). The multi-agent system enhances perception, decision-making, and control capabilities through multimodal interaction interfaces, speech recognition and conversion, visual sensing, and multimodal control. Considering the critical role of trust in system interaction experience, collaboration efficiency, and dynamic strategy adaptation, this study places human-robot trust that integrates operator physiological data, task performance, and interaction experience at its core. The model systematically considers various trust-related factors in XR-supported human-robot assembly and draws on interpersonal trust theory. It enables bidirectional trust quantification, specifically from Agent_R to Agent_H and from Agent_H to Agent_C, moving beyond conventional one-way trust modeling approaches. Furthermore, leveraging XR interaction environments, the study refines the multi-agent collaborative decision-making process and introduces a real-time trust regulation mechanism. Experimental results demonstrate that the proposed XR-driven multi-agent mutual trust evaluation and regulation framework significantly improves collaboration efficiency and system adaptability compared to traditional methods. These findings confirm its strong potential for application in human-robot collaborative assembly systems.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111558"},"PeriodicalIF":6.5,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing the allocation of flood control investments for effective disaster risk reduction","authors":"Kuo-Hao Chang ,&nbsp;Yueh-Ning Hsu ,&nbsp;Hsin-Chi Li","doi":"10.1016/j.cie.2025.111536","DOIUrl":"10.1016/j.cie.2025.111536","url":null,"abstract":"<div><div>Effectively allocating funding for costly flood control infrastructure projects over the long term is critical for mitigating the economic and social disruptions caused by extreme rainfall events, especially as the effects of climate change intensify. In collaboration with the National Science and Technology Center for Disaster Reduction (NCDR) in Taiwan, we develop a novel multi-stage stochastic programming model and a simulation–optimization solution method to optimize the annual allocation of flood control infrastructure investments. The primary objective is to minimize long-term disaster costs under realistic constraints, such as budget limitations, construction schedules, and uncertainties in future rainfall patterns and discount rates. Our approach utilizes an Improved Self-Adaptive Teaching-Learning-Based Optimization (ISATLBO) algorithm to dynamically allocate resources and prioritize infrastructure projects like flood diversion, detention ponds, and river channel widening. The optimization framework is informed by a predictive model for flood costs which leverage up-to-date data, machine learning techniques and the expertise of domain experts. We demonstrate the effectiveness of the proposed methodology through an empirical study in the Sanye Creek Basin in Tainan City, Taiwan. Using a shared budget of 10,000 statistically validated evaluations, ISATLBO cuts expected flood losses by about 5.9% compared with the next-best heuristic (TLBO), outperforming all three established benchmarks. The accompanying sensitivity analysis highlights which economic and climate factors drive risk most strongly, giving decision-makers a clear basis for steering limited annual budgets toward the projects that yield the greatest long-term protection.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111536"},"PeriodicalIF":6.5,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A new model for enhancing collaboration between humanitarian organizations and private sector in humanitarian relief chain","authors":"Iman Shokr,&nbsp;Fariborz Jolai,&nbsp;Ali Bozorgi-Amiri","doi":"10.1016/j.cie.2025.111556","DOIUrl":"10.1016/j.cie.2025.111556","url":null,"abstract":"<div><div>Poor collaboration among organizations results in various challenges to disaster response management. Typically, multiple humanitarian organizations (HOs) are involved in relief chains and need to cooperate with each other, while private sector organizations are expected to improve the performance level of the relief effort. In this paper, we develop a fourth-party humanitarian logistics (4PHL) model to devise a humanitarian relief chain (HRC) with the goal of enhancing the collaboration between multiple HOs and third-party logistics providers (3PLPs) and cope with challenges associated with decentralized (a.k.a. local) planning, a defining characteristic of collaboration in humanitarian logistics. We propose a new two-level two-stage stochastic model to design the 4PHL network, while a robust version of the same model is created as well to better cope with potential uncertainty. An efficient scenario-reduction method is utilized to lower the number of scenarios and streamline the process of solving the model. We implement the model on a real-world case, namely the 2017 Iran-Iraq earthquake, and evaluate the results. In the end, several sensitivity analyses are performed to derive practical insights for policy- and decision-makers.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111556"},"PeriodicalIF":6.5,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FRAM-PSO: A semi-quantitative framework integrating multi-dimensional sustainability criteria","authors":"Ali Karevan,&nbsp;Sylvie Nadeau","doi":"10.1016/j.cie.2025.111560","DOIUrl":"10.1016/j.cie.2025.111560","url":null,"abstract":"<div><div>The increasing complexity of modern industrial systems, particularly those integrating smart wearables, makes it harder for traditional risk analysis methods to keep up. Systemic approaches such as the Functional Resonance Analysis Method (FRAM) help to understand how systems behave; however, there is an opportunity to develop more reliable quantification methods and integrate sustainability criteria, which current methods often do not emphasize. To address these gaps, this paper introduces a novel semi-quantitative framework that integrates FRAM with the Particle Swarm Optimization (PSO). This hybrid approach provides a structured methodology to systematically identify system functions, quantify performance variability, and model risk propagation. A key contribution is the explicit integration of multi-dimensional sustainability criteria (environmental, economic, and social) into the risk management process. This allows for the selection of optimized mitigation strategies. Three case studies involving smart wearables in assembly and disassembly systems were used to demonstrate the effectiveness of the proposed methodology. The results showcase the model’s ability to identify high-risk pathways and prioritize mitigation efforts. This confirms its potential as a decision-support tool. This study contributes a novel methodological structure for embedding sustainability and optimization into systemic risk management.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111560"},"PeriodicalIF":6.5,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nurse-to-patient assignment problem with uncertainties in demand and skill requirements: A stochastic programming approach","authors":"Ngoc-Dai Nguyen ,&nbsp;Nadia Lahrichi ,&nbsp;Chunlong Yu","doi":"10.1016/j.cie.2025.111554","DOIUrl":"10.1016/j.cie.2025.111554","url":null,"abstract":"<div><div>In Home Health Care operations, the Nurse-to-Patient Assignment (NPA) is a critical decision that directly impacts workload balance among nurses, the quality of patient care, and total overtime for home health care providers. The NPA problem involves assigning nurses to patients who require health care services at home over a specified treatment period. Several practical factors must be considered, including the compatibility between a nurse’s skill set and a patient’s needs, continuity of care, and workload balance. However, in real-world scenarios, patient conditions may change over time during the treatment period, potentially requiring different visit frequencies and nurses with varying skill sets. These changes can disrupt continuity of care and lead to potential overtime for nurses. To address this, we formulate the NPA problem while accounting for uncertainty in both patient demand and skill requirements, and propose a stochastic programming approach to optimize nurse-to-patient assignments. The efficiency of the proposed approach and the impact of various problem characteristics are evaluated through extensive computational experiments.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111554"},"PeriodicalIF":6.5,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exact approaches for parallel machine scheduling with loading and unloading servers","authors":"Alessandro Druetto ,&nbsp;Andrea Grosso ,&nbsp;Jully Jeunet ,&nbsp;Fabio Salassa ,&nbsp;Enrico Chiaramello","doi":"10.1016/j.cie.2025.111550","DOIUrl":"10.1016/j.cie.2025.111550","url":null,"abstract":"<div><div>We study the non-preemptive scheduling of general jobs on identical parallel machines, where loading and unloading operations are performed by either a single shared server or by two dedicated servers. The objective is to minimise the makespan. In the literature, exact solution approaches are scarce, and limited to the two-machine case with a single server, or to regular job sets processed on any number of machines with two dedicated servers. We propose three exact solution approaches, each of which is adapted to both problem variants for any number of machines. We adapt two existing models, the time-indexed formulation (TIF) and the arc-flow formulation (FFF), and we propose a constraint programming model (CP). The performance of the models is assessed on benchmark instances with up to 100 jobs and 7 machines. Computational results showed that the FFF and CP models outperformed TIF for instances with more than 10 jobs. CP showed an outstanding performance, solving the majority of problems with 100 jobs to optimality in a short computational time. When non optimal, CP provided feasible solutions with an extremely low gap to best bound.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111550"},"PeriodicalIF":6.5,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-criteria Decision Analysis as a tool for post-processing bias mitigation in machine learning algorithms","authors":"Guilherme Lima Correa,&nbsp;Betania Silva Carneiro Campello,&nbsp;Leonardo Tomazeli Duarte","doi":"10.1016/j.cie.2025.111552","DOIUrl":"10.1016/j.cie.2025.111552","url":null,"abstract":"<div><div>This study proposes a novel post-processing bias mitigation strategy using Multicriteria Decision Analysis (MCDA) to evaluate machine learning (ML) models based on fairness and performance metrics. As ML becomes increasingly integrated into decision-making systems, concerns have emerged about unfair treatment of protected groups defined by attributes such as race, gender, or age. Fair ML techniques aim to address this issue through methods that mitigate bias at different stages of model development. Our approach operates in the post-processing stage, without requiring any modifications to the training process, data flow, or underlying algorithm. We apply MCDA to assess seven widely-used ML algorithms using real benchmark datasets and multiple fairness and accuracy criteria. The results show that MCDA enables a transparent evaluation of trade-offs between fairness and performance. Furthermore, our method is easy to integrate into existing workflows and supports fairer model selection without the need for retraining, offering practical value for decision-makers.</div></div>","PeriodicalId":55220,"journal":{"name":"Computers & Industrial Engineering","volume":"210 ","pages":"Article 111552"},"PeriodicalIF":6.5,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}