Article
Operations Research & Management Science
Martin Pouls, Nitin Ahuja, Katharina Glock, Anne Meyer
Summary: This work presents a forecast-driven idle vehicle repositioning algorithm for dynamic ride-sharing systems. By maximizing trip request acceptance rates and minimizing vehicle travel times, this algorithm improves overall performance. Extensive simulation studies show that it significantly reduces trip request rejection rates and enhances customer waiting and ride times compared to benchmark algorithms.
ANNALS OF OPERATIONS RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Jianguo Chen, Kenli Li, Keqin Li, Philip S. Yu, Zeng Zeng
Summary: The article proposes a BSDP system to dynamically provide the optimal bicycle station layout for the DL-PBS network, consisting of four key modules: bicycle drop-off location clustering, bicycle-station graph modeling, bicycle-station location prediction, and bicycle-station layout recommendation. Experiments demonstrate the effectiveness, accuracy, and feasibility of the system.
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2021)
Article
Computer Science, Interdisciplinary Applications
Shi-Zhi Chen, De-Cheng Feng
Summary: This paper discusses a machine learning approach based on multifidelity data to improve the accuracy of prediction models in structural behavior. The feasibility of this method is validated through a case study, and the impact of different factors on model performance is thoroughly investigated.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2022)
Article
Energy & Fuels
Ziteng Huang, Ran Li, Zhangxin Chen
Summary: Steam-assisted gravity drainage (SAGD) is a widely used thermal recovery process in oil sands production. The traditional numerical simulation process for SAGD prediction is time-consuming and computationally expensive. This study proposes a data-driven model-based workflow that utilizes machine learning algorithms to predict future SAGD production performance based on historical production data and operational conditions. Comparison of different algorithms reveals that the GRU-based model has the best predictive ability.
Article
Chemistry, Multidisciplinary
Shuo Zhu, Xianzhi Song, Zhaopeng Zhu, Xuezhe Yao, Muchen Liu
Summary: This study presents three methods for predicting stuck pipe, including detection of friction coefficient, prediction of stuck pipe probability using artificial neural network, and establishment of a comprehensive indicator using fuzzy mathematics. The results indicate that the last model is the best, with a high prediction accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Civil
Md Mintu Miah, Stephen P. Mattingly, Kate Kyung Hyun, Joseph Broach, Nathan Mcneil, Sirisha Kothuri
Summary: This study aims to determine the best buffer types and sizes and evaluate the impact of local characteristics on buffer type and size selection. The results recommend that a generalized model with the combination of Network and Euclidean buffers of multiple sizes provides the best prediction of AADBT, and Network buffers outperform Euclidean buffers. However, city-specific models with a single type and size of buffer sometimes outperform the generalized model.
Article
Construction & Building Technology
Lavinia Chiara Tagliabue, Fulvio Re Cecconi, Stefano Rinaldi, Angelo Luigi Camillo Ciribini
Summary: The concept of the built environment as a container for human activities has shifted towards focusing on the user's well-being and experience, especially in educational facilities, where indoor air quality significantly impacts learning performance.
ENERGY AND BUILDINGS
(2021)
Article
Biochemical Research Methods
Loic Chagot, Cesar Quilodran-Casas, Maria Kalli, Nina M. Kovalchuk, Mark J. H. Simmons, Omar K. Matar, Rossella Arcucci, Panagiota Angeli
Summary: The control of droplet formation and size using microfluidic devices is essential for laboratory and industrial applications. This study investigates how surfactants can improve stability and control droplet size in silicone oil. Two data-driven models, based on Bayesian regularised artificial neural network and XGBoost, accurately predict droplet size as a function of flow rates and surfactant properties. Compared to semi-empirical models, the data-driven models have higher accuracy. The models were also trained with simplified inputs, showing good prediction results. Additionally, a synthetic data set was generated, proving the feasibility of using this method in future lab on a chip applications.
Article
Computer Science, Interdisciplinary Applications
Huan Luo, Stephanie German Paal
Summary: This paper proposes a novel computational method to improve the accuracy and efficiency of dynamic response computation. It models the parameters of the system by minimizing the objective function and bypasses the need for training data and time step requirements. It also develops an efficient step-by-step solver and analyzes the extension of this method for nonlinear dynamic response computation.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2023)
Article
Energy & Fuels
Laisuo Su, Mengchen Wu, Zhe Li, Jianbo Zhang
Summary: This study demonstrates the capability of machine learning techniques to accurately predict the cycle life of lithium-ion batteries by capturing hidden features in complex, nonlinear systems.
Article
Energy & Fuels
Yunlu Li, Guiqing Ma, Junyou Yang, Yan Xu
Summary: In the task of system analysis for VSG cluster, the widely used aggregation modeling method often leads to inevitable errors. To improve the accuracy, a data-physical driven modeling method is proposed, which includes a hybrid model structure consisting of single machine aggregation model and deep neural network based aggregated-error model. By analyzing the equivalence between aggregation error and black box modeling issue, the proposed method is validated through test cases under large disturbance and multi-operating points conditions, demonstrating satisfactory modeling accuracy.
Article
Computer Science, Artificial Intelligence
Hengyu Man, Xiaopeng Fan, Ruiqin Xiong, Debin Zhao
Summary: This paper introduces a novel data clustering-driven neural network (TreeNet) for intra prediction in video compression. TreeNet constructs networks in a tree-structured manner and clusters the training data to improve prediction performance. Experimental results show that TreeNet can save an average of 3.78% bitrate when assisting VVC intra modes.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Lining Wang, Tien Ju Lee, Jan Bavendiek, Lutz Eckstein
Summary: Anthropometry is a science used in various disciplines, particularly in apparel and product design, to measure human body dimensions. The study introduces a neural network architecture that accurately predicts detailed body measurements, outperforming existing regression models.
APPLIED SOFT COMPUTING
(2021)
Article
Telecommunications
Amir Ghasemi, Janaki Parekh
Summary: The rapid development of wireless technologies in the past decade has led to increased pressure on limited radio spectrum resources. To improve allocation efficiency, regulators are considering dynamic spectrum sharing. We propose DeepAir, a robust and scalable model that can learn and predict complex temporal and spectral dependencies in multivariate spectrum data. With test results showing high accuracy and applicability to different sensors, DeepAir can enhance spectrum sharing capabilities.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2023)
Article
Computer Science, Information Systems
Xueli Wang, Shangwei Zhao, Xin Wang, Ming Yang, Xiaoming Wu
Summary: This study investigates the data-driven control problem for discrete-time nonlinear systems with state constraints and dynamic quantization. A dynamic quantizer is used to encode the control signal and alleviate quantization errors. By employing three sets of neural networks, a data-based goal representation heuristic dynamic programming algorithm is developed to achieve near-optimal control without requiring a model. The proposed control strategy includes a barrier-based control strategy for generating repulsive control force at constraint boundaries, a new weight updating rule based on historical data, and demonstration of bounded weight estimation errors using the Lyapunov method.
INFORMATION SCIENCES
(2023)
Article
Construction & Building Technology
T. D. Toan, N. H. Long, Y. D. Wong, T. Nguyen
Summary: This study investigates the effect of variability in layer thickness and elastic modulus of pavement materials on the reliability of flexible pavement structural performance. Monte Carlo simulation was used to establish the probability distribution of reliability. It was found that changes in coefficient of variation (COV) affect the reliability of flexible pavement, with upper layer thickness and elastic modulus having a greater impact on reliability.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(2023)
Article
Engineering, Civil
Kum Fai Yuen, Ling Qian Choo, Xue Li, Yiik Diew Wong, Fei Ma, Xueqin Wang
Summary: This study analyzes the determinants of user acceptance of fully autonomous public transport from multiple theoretical perspectives, finding that users' value perception of APT significantly influences their acceptance of it.
Article
Management
Xue Li, Kum Fai Yuen, Xueqin Wang, Yiik Diew Wong
Summary: The ongoing COVID-19 pandemic has led to a surge of interest in contactless technology applications. This study investigates the psychological determinants of emerging contactless technologies by synthesising theoretical insights from the health belief model and technology acceptance model. The results show that health belief factors and technological characteristics contribute to the willingness to adopt contactless technologies.
TECHNOLOGY ANALYSIS & STRATEGIC MANAGEMENT
(2023)
Article
Green & Sustainable Science & Technology
Maohao Che, Yiik Diew Wong, Kit Meng Lum, Maria Cecilia Rojas Lopez
Summary: The intention-behavior relationship is examined in a case study of pedestrians and cyclists on a shared footpath in Singapore. The findings suggest that keep left markings are effective in improving users' behavioral intention, but pedestrian habits make it difficult for them to change their behavior. The study also found no significant changes in pedestrian-cyclist interaction patterns or perceived conflict level and safety level after the keep left treatment.
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION
(2023)
Article
Green & Sustainable Science & Technology
Jie Zhang, Meng Meng, David Z. W. Wang, Li Zhou, Linghui Han
Summary: This paper investigates the problem of bike allocation in a competitive bike sharing market. A continuum approximation (CA) approach is used to handle computational challenges by assuming that allocation points and user demand are continuously distributed in a two-dimensional region. Bike sharing companies bear allocation and bike depreciation costs while earning revenue from fare collection. User's choice of bike service depends on walking distance and bike quality preference. The demand elasticity is considered in relation to the density of allocation points. A leader-follower Stackelberg competition model is developed to derive the optimal allocation strategy for the market leader. Numerical studies are conducted for both hypothetical and real cases to examine the impact of parameters on model performance and demonstrate the application of the proposed model in decision making.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Transportation Science & Technology
Duowei Li, Feng Zhu, Tianyi Chen, Yiik Diew Wong, Chunli Zhu, Jianping Wu
Summary: This study proposes a hierarchical control model, COOR-PLT, that uses deep reinforcement learning to coordinate adaptive platoons of connected and autonomous vehicles (CAVs) at signal-free intersections. The model combines a centralized control strategy to form adaptive platoons and a decentralized control strategy to coordinate multiple platoons passing through the intersection. Simulation results show that the model achieves satisfactory convergence, adapts platoon sizes based on traffic conditions, and avoids deadlocks at the intersection. Compared to other control methods, the model demonstrates superiority in adopting adaptive platooning and deep reinforcement learning-based coordination strategies, resulting in reduced travel time and fuel consumption in different traffic conditions.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Management
Xueqin Wang, Yiik Diew Wong, Wenming Shi, Kum Fai Yuen
Summary: This study investigates consumers' preferences for parcel delivery by considering their contributions in co-creating delivery services. The findings suggest that consumers who choose alternative delivery options are more willing to contribute physical effort, but less interested in responding attentively to informational updates. Efforts required for social interactions discourage consumers from choosing attended deliveries, prompting unattended alternatives as more attractive choices.
INTERNATIONAL JOURNAL OF LOGISTICS MANAGEMENT
(2023)
Article
Transportation
Trinh Dinh Toan, Soi Hoi Lam, Meng Meng, Yiik Diew Wong
Summary: This research study explores travel demand management (TDM) strategies in Singapore and its potential transferability to Hanoi, Vietnam. The study critically analyzes international policies and conducts a case study of Singapore to understand policy transfer initiatives in TDM. It finds considerable differences between Singapore and Hanoi in terms of development objectives, social and political settings, and governance structure, suggesting constraints on policy transfer. However, the findings have valuable policy implications for Hanoi and provide insights for developing a roadmap for TDM policy transfer.
CASE STUDIES ON TRANSPORT POLICY
(2023)
Article
Polymer Science
Jun Yang, Xingyu Yi, Huimin Chen, Yiik Diew Wong, Yulou Fan, Wei Huang
Summary: The utilization of reclaimed asphalt pavement can reduce the cost of pavements containing epoxy polymer. This study aimed at improving the homogeneity of an EP-reclaimed asphalt mixtures at both the micro- and meso-scale and found that mixing temperature and reclaimed asphalt gradation have significant effects on the material properties.
Article
Economics
Shanshan Sun, Yiik Diew Wong
Summary: Limits in attention during driving hinder drivers from engaging in non-driving activities, but autonomous vehicles offer a solution. Through interviews with 763 car drivers, we constructed an integrated choice model and found that drivers' multitasking habits discourage their adoption of autonomous vehicle services, while their addiction to smart devices encourages it.
Article
Transportation
Shanshan Sun, Yiik Diew Wong, Xueqin Wang, Andreas Rau
Summary: This study examines the causality of travel-based multitasking behavior using three theoretical frameworks. The results show that habit has the strongest impact on multitasking behavior, and norm significantly affects habit formation and smart device addiction. Policy-makers should consider the differences among intention, habit, and addiction in designing interventions.
TRAVEL BEHAVIOUR AND SOCIETY
(2024)
Review
Business
Bohao Ma, Yiik Diew Wong, Chee-Chong Teo, Ziyan Wang
Summary: This research uses large-scale user-generated content to decipher consumers' quality perceptions in the online food delivery sector. By applying machine learning algorithms, key service topics related to the consumer experience are identified. Ultimately, our findings provide crucial theoretical and practical implications for practitioners and researchers in this field.
JOURNAL OF RETAILING AND CONSUMER SERVICES
(2024)
Article
Construction & Building Technology
Xingyu Yi, Huimin Chen, Yiik Diew Wong, Jun Yang, Wei Huang
Summary: The reutilization of reclaimed asphalt pavement (RAP) is important for sustainable environmental policies. This study investigated the effects of epoxy polymer (EP) content on the properties of recycled asphalt binder. The addition of EP significantly enhanced the strength and resistance of the binder but reduced its elongation.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Ergonomics
Bo Wang, Tianyi Chen, Chi Zhang, Yiik Diew Wong, Hong Zhang, Yunhao Zhou
Summary: This study proposes an improved safety potential field, called the Work-Zone Crash Risk Field (WCRF), to measure the crash risk in transition areas of highway work zones. The WCRF-based surrogate safety measure (SSM) outperforms conventional SSMs and offers a practical and comprehensive way to describe the crash risk in work zones. The developed WCRF technique allows for identifying key risk-contributing features, facilitating the development of safety management strategies for work zones.
ACCIDENT ANALYSIS AND PREVENTION
(2024)
Article
Engineering, Industrial
Hiroshi Matsuhisa, Nobuo Matsubayashi
Summary: This study investigates the formation of an alliance between competing manufacturers and a monopolistic platform retailer, and analyzes the impact of the degree of differentiation among manufacturers on the formation of the alliance and the profitability of the retailer.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Lingxuan Kong, Ge Zheng, Alexandra Brintrup
Summary: Supply Chain Financing is used to optimize cash flows in supply networks, but recent scandals have shown inefficiencies in risk evaluation. This paper proposes a Federated Learning framework to address order-level risk evaluation.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Jing Gu, Xinyu Shi, Junyao Wang, Xun Xu
Summary: The asymmetric market power between a firm and its partners negatively affects the firm's financial performance. Building relationships with suppliers or customers that have matched market power is the best approach. The strength of the buyer-supplier relationship amplifies the negative impact of asymmetric market power, while the level of relationship embeddedness reduces its negative effect. Moreover, firm-specific institutional, industry, and regional economic heterogeneities also influence the financial impact of asymmetric market power.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Yu Du, Jun-qing Li
Summary: This study investigates the group scheduling of a distributed flexible job shop problem using the concrete precast process. The proposed solution utilizes three coordinated double deep Q-networks (DQN) as a learn-to-improve reinforcement learning approach. The algorithm shows superiority in minimizing costs and energy consumption.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Xiaoyu Yan, Weihua Liu, Ou Tang, Jiahe Hou
Summary: This study analyzes the market amplification effect and the impact of entrant's overconfidence on a two-sided platform. The results show that overconfident entrants can lead to price increases and benefit both the existing firms and themselves to a certain extent.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Illya Kaynov, Marijn van Knippenberg, Vlado Menkovski, Albert van Breemen, Willem van Jaarsveld
Summary: The One-Warehouse Multi-Retailer (OWMR) system is a typical distribution and inventory system. Previous research has focused on heuristic reordering and allocation strategies, which are time-consuming and problem-specific. This paper proposes a Deep Reinforcement Learning (DRL) algorithm for OWMR problems, which infers a multi-discrete action distribution and improves performance with a random rationing policy.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Yimeng Sun, Ruozhen Qiu, Minghe Sun
Summary: This study considers a multi-period inventory management problem for a retailer offering limited-time discounts and having a joint service-level requirement under demand uncertainty. It proposes a double-layer iterative approach to solve the problem and maximize total profit while balancing the service level using a posteriori method and an affinely adjustable robust chance-constrained model.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Anas Neumann, Adnene Hajji, Monia Rekik, Robert Pellerin
Summary: This paper presents a new mathematical formulation for planning and scheduling activities of Engineer-To-Order (ETO) projects, along with a new ETO strategy to reduce the impacts of design uncertainty. The study proposes a hybrid Layered Genetic Algorithm combined with an adaptive Lamarckian learning process (LLGA) and compares it with the branch-and-cut procedure of CPLEX. The results show good performance of the proposed mathematical model for small and medium-sized instances.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Thilini Ranasinghe, Chanaka D. Senanayake, Eric H. Grosse
Summary: Production systems are undergoing transformative changes, necessitating adaptability from human workers. This study developed an analytical model to account for stochastic processing times and learning heterogeneity, revealing insights into system performance.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Sunil Tiwari, Pankaj Sharma, Ashish Kumar Jha
Summary: Black Swan events such as the COVID-19 pandemic and the Suez Canal blockage have a significant impact on firms' technology adoption decisions, especially in terms of disruptions and digitalization in the supply chains. This study investigates the influence of institutional forces and environmental contingencies on supply chain digitalization from an institutional and contingency theory perspective. The findings emphasize the importance of organizational readiness and people readiness, including top management involvement and employee training, in facilitating digitalization.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Fabio Neves-Moreira, Pedro Amorim
Summary: Omnichannel retailers are using stores as distribution centers to provide faster online order fulfillment services. However, in-store picking operations can impact the offline customer experience. To address this, we propose a Dynamic In-store Picker Routing Problem (diPRP) that minimizes customer encounters while fulfilling online orders. Our solution approach combines mathematical programming and reinforcement learning to find efficient picking policies that reduce customer encounters.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Article
Engineering, Industrial
Richard Kraude, Ram Narasimhan
Summary: In this study, the relationship between Vertical Integration (VI) and Environmental Performance (EP) is examined, revealing that highly integrated firms produce less waste but engage in fewer environmental initiatives. These findings are crucial for understanding the impact of stakeholder exposure on organizational behavior.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)
Review
Engineering, Industrial
Korina Katsaliaki, Sameer Kumar, Vasilis Loulos
Summary: This research conducts a systematic literature review (SLR) and content analysis on Supply Chain Coopetition (SCC) through the PRISMA framework. It examines the theory of coopetition and organizational relationships in intra-firm and inter-firm supply chains, focusing on collaboration between rival manufacturers. The study identifies structures and mechanisms of coopetition, such as buyer-supplier coopetition, supply networks coopetition, and production and distribution/logistics coopetition. It provides a holistic approach to SCC management practices and serves as a guide for future research.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2024)