Transportation Research Part E-Logistics and Transportation Review最新文献

{"title":"Personalized origin–destination travel time estimation with active adversarial inverse reinforcement learning and Transformer","authors":"Shan Liu ,&nbsp;Ya Zhang ,&nbsp;Zhengli Wang ,&nbsp;Xiang Liu ,&nbsp;Hai Yang","doi":"10.1016/j.tre.2024.103839","DOIUrl":"10.1016/j.tre.2024.103839","url":null,"abstract":"<div><div>Travel time estimation is important for instant delivery, vehicle routing, and ride-hailing. Most studies estimate the travel time of specified routes, and only a few studies pay attention to origin–destination travel time estimation (OD-TTE) without a specified route. Moreover, most of these studies on OD-TTE ignore the personalized route preference and the cost of data annotation. To fill this research gap, we analyze the individual route preference and propose a personalized origin–destination travel time estimation method based on active adversarial inverse reinforcement learning (AA-IRL) and Transformer. To analyze the personalized route preference, we integrate adversarial inverse reinforcement learning with active learning, which effectively reduces the cost of sample annotation. After inferring the possible routes, we propose AdaBoost multi-fusion graph convolutional Transformer network (AMGC-Transformer) for travel time estimation. Numerical experiments conducted on ride-hailing and online food delivery trajectories in China validate the advantage of our method. Compared to relevant studies, our approach can improve F1-score of route inference by 2.50–3.35% and reduce the mean absolute error of OD-TTE by 7.44–11.66%.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103839"},"PeriodicalIF":8.3,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142571981","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":"Dynamic matching radius decision model for on-demand ride services: A deep multi-task learning approach","authors":"Taijie Chen ,&nbsp;Zijian Shen ,&nbsp;Siyuan Feng ,&nbsp;Linchuan Yang ,&nbsp;Jintao Ke","doi":"10.1016/j.tre.2024.103822","DOIUrl":"10.1016/j.tre.2024.103822","url":null,"abstract":"<div><div>As ride-hailing services have experienced significant growth, most research has concentrated on the dispatching mode, where drivers must accept the platform’s assigned trip requests. However, the broadcasting mode, in which drivers can freely choose their preferred orders from those broadcast by the platform, has received less attention. One crucial but challenging task in such a system is the determination of the matching radius, which usually varies across space, time, and real-time supply/demand characteristics. This study develops a <strong>D</strong>eep <strong>L</strong>earning-based <strong>M</strong>atching <strong>R</strong>adius <strong>D</strong>ecision (DL-MRD) model that predicts key system performance metrics for a range of matching radii, which enables the ride-hailing platform to select an optimal matching radius that maximizes overall system performance according to real-time supply and demand information. To simultaneously maximize multiple system performance metrics for matching radius determination, we devise a novel multi-task learning algorithm named <strong>W</strong>eighted <strong>E</strong>xponential <strong>S</strong>moothing <strong>M</strong>ulti-task (WESM) learning strategy that enhances convergence speed of each task (corresponding to the optimization of one metric) and delivers more accurate overall predictions. We evaluate our methods in a simulation environment designed for broadcasting-mode-based ride-hailing service. Our findings reveal that dynamically adjusting matching radii based on our proposed approach significantly improves system performance.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103822"},"PeriodicalIF":8.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142571980","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":"Stockyard allocation in dry bulk ports considering resource consumption reduction of spraying operations","authors":"Wenyuan Wang ,&nbsp;Jiaqi Guo ,&nbsp;Qi Tian ,&nbsp;Yun Peng ,&nbsp;Zhen Cao ,&nbsp;Keke Liu ,&nbsp;Shitao Peng","doi":"10.1016/j.tre.2024.103816","DOIUrl":"10.1016/j.tre.2024.103816","url":null,"abstract":"<div><div>Stockyard allocation is a crucial segment of operational decision-making in dry bulk ports (DBPs). The stockyard allocation plan determines the storage position and duration of each stockpile to avoid operational delays in stockyards. Spraying operations, a unique operation in DBPs, are significantly influenced by stockyard allocation plans. Port operators regularly conduct spraying operations to prevent dust diffusion during the storage of dry bulk materials in stockyards. The spraying operation system consumes substantial electrical energy to transport the water to the designated material pile and spray large amounts of water onto its surface. Due to the layout constraints of pipelines and spraying nozzles, different stockyard allocation plans lead the varying consumptions of electrical energy and water resources for spraying operations. However, previous studies on the stockyard allocation problem frequently ignore the impacts of the stockyard allocation plan on the resource consumption of spraying operations. To fill this gap, this paper proposes a stockyard allocation model that uniquely considers the resource consumption of spraying operations to balance operation efficiency and resource consumption in stockyards from a global perspective. With the goal of minimizing the total cost, including operation delay penalties in stockyards and the electricity and water costs of spraying operations, a series of comprehensive experiments was conducted based on practical data collected from a major DBP in China under varying stockpile densities and stockyard efficiency properties. The results clearly show significant differences in the stockyard allocation plan and the total cost resulting from considering and disregarding the resource consumption of spraying operations in the stockyard allocation decision-making process. With only a 3.09% increase in average delay time in stockyards, the proposed model can reduce the total cost by 19.26%, the electricity cost by 54.06% and the water cost reduction by 35.09%. Meanwhile, the carbon emissions are reduced 75 tons on average for spraying operations and the Whale Optimization Algorithm (WOA) performs well on large-scale instances. The proposed model can avoid unnecessary resource consumption of spraying operations with acceptable operation delay penalties in stockyards.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103816"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553679","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":"Inhibitors in ridesharing firms from developing Nations: A novel Integrated MCDM – Text Mining approach using Large-Scale data","authors":"Souradeep Koley ,&nbsp;Mukesh Kumar Barua ,&nbsp;Arnab Bisi","doi":"10.1016/j.tre.2024.103832","DOIUrl":"10.1016/j.tre.2024.103832","url":null,"abstract":"<div><div>Our study identifies major impediments (or inhibitors) faced by Transportation Network Companies (TNCs) such as Uber, Lyft, and Ola within the context of developing nations. While existing studies on TNCs centered on passenger adoption and drivers’ perspectives, we quantitively assess the inhibitors and provide mitigation strategies. To achieve this, we use machine learning methods, particularly Latent Dirichlet Allocation (LDA) and emotion analysis on large-scale public data, to understand and classify consumer perspectives on TNCs into multiple themes. The latent theme helps experts of different ridesharing firms get a holistic perspective of riders on TNCs, assisting them in identifying the inhibitors. Using the Delphi method, we were able to achieve a consensus in identifying six primary and nineteen secondary inhibitors. We rank the primary inhibitors based on the optimal weight obtained using the Bayesian Best Worst Method. To minimize uncertainty and imprecise judgment in decision-making, we combine the grey theory with the Decision-Making Trial and Evaluation Laboratory (Grey-DEMATEL) to identify the interrelationships among the secondary inhibitors. Moreover, we perform sensitivity analysis to show the robustness of our solution. Contrary to conventional perception, our findings indicate that the government is the primary inhibitor for TNCs due to current policy and discrepancies in regulations between central and states. Additionally, our studies introduce five new inhibitors to the literature, which include drivers inciting trip cancellation to avoid commission, internal coalition of drivers, commission miscomprehension among drivers, limited infrastructure for cashless operation, and internal conflict and dysfunction within the department. The findings from large-scale data analysis, coupled with group decision-making, offer various managerial implications that can guide future managers and policymakers to enhance the operational efficiency of firms.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103832"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553663","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":"Socially responsible e-commerce supply chains: Sales mode preference and store brand introduction","authors":"Xinxin Zhang ,&nbsp;Xiuyi Zhang ,&nbsp;Junran Huang","doi":"10.1016/j.tre.2024.103829","DOIUrl":"10.1016/j.tre.2024.103829","url":null,"abstract":"<div><div>Motivated by the widespread adoption of corporate social responsibility (CSR), we investigate a socially responsible e-commerce supply chain where the E-platform owns a store brand product and supports online sales of the manufacturer’s product under agency selling or reselling. The socially responsible firm has a mixed objective of its profit and consumer surplus. We explore how the firms’ CSR concern affects their decisions and economic performance. Our results contradict conventional wisdom which suggests that a firm has to sacrifice profitability to achieve social responsibility and that a firm’s CSR concern benefits its supply chain partner. Instead, we show that under agency selling, the E-platform’s concern on consumers can improve its own profit while harming the manufacturer’s profit. Furthermore, when both firms are socially responsible, their consumer concern can improve their profits simultaneously under reselling, leading to a “doing well by doing good” effect. As horizontal or vertical differentiation between the two products increases, this effect is more likely to be realized. Regarding firms’ sales mode preferences, in a traditional for-profit supply chain, agency selling is the only mode preferred by both parties. However, in a socially responsible supply chain, they can achieve preference alignment under either agency selling or reselling.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103829"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553680","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":"Distributionally robust optimization for minimizing price fluctuations in quota system","authors":"Chi Xie ,&nbsp;Zheng Cui ,&nbsp;Daniel Zhuoyu Long ,&nbsp;Jin Qi","doi":"10.1016/j.tre.2024.103812","DOIUrl":"10.1016/j.tre.2024.103812","url":null,"abstract":"<div><div>Quota systems play a crucial role in regulating public-interest goods and controlling negative externalities, with a primary focus on social impacts rather than economic benefits. This paper examines the decision-making process for quota release, aiming to control growth rates and ensure price stability over time. We first develop a chance-constrained problem for quota systems, solving it using sample average approximation. Due to computational demands, alternative approximation methods are explored. We consider two types of quota systems: mature systems with known distributions and newly established systems with distributional ambiguity. For mature systems, Conditional Value-at-Risk (CVaR) is used to approximate the chance constraint, while for newly established systems, worst-case CVaR is employed within a robust optimization framework and the binary search algorithm is derived to efficiently solve the problem. The proposed models’ effectiveness is validated through computational studies using data from Singapore’s Vehicle Quota System. With known distributions, our CVaR sample average approximation (CVaR-SAA) model outperforms traditional models, reducing violation probability by more than 56.32%. With distributional ambiguity, worst-case CVaR approximation robust optimization (WCVaR-RO) model provides superior solutions, particularly in maximum violation probability (MVP). In the most notable case, WCVaR-RO reduces the MVP by over 53.37%. This research offers valuable insights into the management of quota systems.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103812"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553677","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":"The effect of geographic risk factors on disaster mass evacuation strategies: A smart hybrid optimization","authors":"Ahmad Jafarian ,&nbsp;Tobias Andersson Granberg ,&nbsp;Reza Zanjirani Farahani","doi":"10.1016/j.tre.2024.103825","DOIUrl":"10.1016/j.tre.2024.103825","url":null,"abstract":"<div><div>This paper investigates an urban Emergency Evacuation Network Design (EEND) problem on a large scale when geographical risk in different areas varies. The decisions to make are (i) determining active shelters, (ii) selecting evacuation routes, and (iii) managing the supply of relief commodities from distribution centers to shelters. A region prone to floods and hurricanes is divided into zones, each with a specific vulnerability risk. For each zone, a risk measure is calculated by combining the risk factors –transporting people and relief commodities and the placement of temporary shelters. The objective is to minimize the maximum risk across the network, ensuring a balanced distribution of risk. A combinatorial scenario planning approach is developed to manage the uncertainty in disaster severity and the evacuee numbers. To incorporate varied geographical risks, a smart hybrid optimization approach as a new solution technique is developed, tuned, and validated to solve the EEND problem. The proposed approach uses directed local search structures designed for the EEND problem and an AI-based self-parameter tuning module, enhancing performance. To extract insights, Rennes, France, is considered a case study. The results indicate a reduction in casualties using a min–max formulation compared to traditional sum-risk objectives. Further, a detailed evacuation plan that increases the number of city regions enhances EEND performance. Practical insights suggest minimizing the number of shelters to the essential capacity needed to host all evacuees, as additional shelters may lead to increased evacuation and supply routes, potentially in areas with higher risk.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103825"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553676","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":"Probabilistic forecast-based procurement in seaborne forward freight markets under demand and price uncertainty","authors":"Burakhan Sel ,&nbsp;Stefan Minner","doi":"10.1016/j.tre.2024.103830","DOIUrl":"10.1016/j.tre.2024.103830","url":null,"abstract":"<div><div>Volatility in freight rates and shipping demand poses financial risks for charterers and ship owners. Freight forward agreements (FFAs) are popular hedging tools for fixing freight rates in advance by specifying the amount of cargo to be transported at the maturity period of the agreement. Procurement decisions with FFAs require assessing future freight rates and shipping demand. Accepting an FFA price offer higher than future FFA and spot prices or procuring a larger amount than the actual demand constitutes risks for charterers. We consider the freight procurement problem of a charterer, minimizing the total expected cost using FFA and spot markets under price and demand uncertainty. We show that a state-dependent base-stock policy is optimal with non-decreasing base-stock levels as the demand period approaches when price and demand forecasts are not updated. To determine base-stock levels, we propose probabilistic forecast-based policies with updated forecasts and an increasing base-stock level policy (IBP) adjusting base-stock levels based on the number of periods left until the demand period. The proposed methods are compared with benchmark methods using synthetic data covering different market conditions and real data from 14 bulk and tanker routes. The evaluation period covers pre-crisis (2016–2019) and during-crisis periods (2020–2023), considering major events after 2019, such as the COVID-19 pandemic and the Russia–Ukraine conflict, which led to high market volatility. Numerical evaluations show that policies based on probabilistic forecasts outperform those based on point forecasts. Utilizing probabilistic demand forecasts results in lower costs than probabilistic price forecasts. Experiments on the market data show that IBP results in the lowest cost on average while avoiding excessive procurement due to being in line with the optimal procurement policy. IBP outperforms probabilistic forecast-based policies due to forecast biases in the volatile freight market.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103830"},"PeriodicalIF":8.3,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142553678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Resilience enhancement of multi-modal public transportation system via electric bus network redesign","authors":"Zhongshan Liu ,&nbsp;Bin Yu ,&nbsp;Li Zhang ,&nbsp;Yuxuan Sun","doi":"10.1016/j.tre.2024.103810","DOIUrl":"10.1016/j.tre.2024.103810","url":null,"abstract":"<div><div>The multi-modal public transportation system incorporating the electric bus network and the metro network plays a crucial role in meeting the daily transportation demands in urban areas. However, the inadequate connectivity between the electric bus network and the metro network has resulted in poor resilience of the multi-modal public transportation system. When disruptions occur at metro links or metro stations, stranded passengers cannot be rapidly evacuated through bus lines. Therefore, operators need to redesign the multi-modal public transportation system to enhance the integration between the electric bus network and the metro network. Since it is challenging to modify the fixed metro network, we thus focus on introducing redesign plans for the electric bus network, enabling the multi-modal public transportation system to exhibit the desired resilience in scenarios of disruptions. This paper proposes a two-level framework integrating electric bus network redesign at the tactical level and resource deployment at the planning level. For the redesign problem at the tactical level, this paper designs a tailored branch-and-price algorithm to generate high-quality redesign solutions for the electric bus network. For the resource deployment problem at the planning level, this paper determines the locations of charging facilities and the number of electric buses based on the redesigned electric bus network, considering uncertain passenger demands and metro capacities. We propose a two-layer robust optimization model for the resource deployment problem and develop a tailored column-and-constraint generation algorithm to solve it. Finally, this paper tests the performance of the developed models and algorithms on a set of instances in Beijing. The impact of the uncertainty budget, the number of electric buses, bus capacity, and charging time of electric buses on the system performance is discussed.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103810"},"PeriodicalIF":8.3,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142539169","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":"The role of drone technology and application of IoT on vaccine supply chain during a pandemic under uncertain Environment: A real case study of COVID-19 in Iran","authors":"Nadia Ansari ,&nbsp;Parviz Fattahi ,&nbsp;Mahdyeh Shiri","doi":"10.1016/j.tre.2024.103831","DOIUrl":"10.1016/j.tre.2024.103831","url":null,"abstract":"<div><div>Vaccination is a crucial way to combat the pandemic; in other words, vaccines play an important role in controlling the spread of the virus and ultimately ending the pandemic. This study presents a multi-objective mixed integer linear programming model for the vaccine supply chain considering uncertain cost, vaccine purchase, and lead time. Through the utilization of Internet of Things technology, data about various groups is collected. Upon identification of individuals with good health, the specific needs of each area are ascertained during each period. Subsequently, a mathematical model for the vaccine supply chain is presented, encompassing four distinct levels; manufacturers, distribution centers, health centers, and immunization centers. Furthermore, this model incorporates the utilization of drones to deliver vaccines from distribution centers to health centers because of the significant distance between these two levels. The proposed framework encompasses two main goals; minimizing the total cost and the waiting time for people in the queue. A novel fuzzy approach has been employed to deal with the uncertain parameters. The model’s validation is accomplished through the implementation of a real case study of COVID-19 in Iran. The findings indicate that the lack of Internet of Things technology implementation results in a higher number of individuals being directed to immunization centers, thereby elevating the likelihood of infection, and, this scenario leads to the unnecessary administration of vaccines, leading to resource wastage. Additionally, without using drones, vaccines cannot be delivered and injected into people on time. Ultimately, the proposed framework and methodology can be applied in almost larger dimensions and the results demonstrate the model and methods’ efficiency and effectiveness. Since this study is applied to the case study of COVID-19, the findings can be applied in the conditions of similar pandemics.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"193 ","pages":"Article 103831"},"PeriodicalIF":8.3,"publicationDate":"2024-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142532249","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}