CS Events

PhD Defense Balancing Efficiency and Experience: A Predictive Cyber Physical System (CPS) for Urban Logistics
Export Event Save iCal Send to Google Calendar Send to Yahoo Calendar Send to Outlook Live Send to Microsoft Outlook Save iCal Preserve formatting in description (only supported in some calendar applications) Close Monday, September 23, 2024, 09:00am - 10:30am

Speaker: Shuxin Zhong

Location : CoRE 305

Committee: 

Prof.Desheng Zhang (Advisor), Rutgers CS Graduate Faculty

Prof.Yongfeng Zhang, Rutgers CS Graduate Faculty

Prof.Dong Deng, Rutgers CS Graduate Faculty

Prof.Fei Miao, University of Connecticut, Pratt & Whitney Endowed Associate Professor

Event Type: PhD Defense

Abstract: Cyber-Physical Systems (CPS) integrate physical entities with information systems, enabling sensing, decision-making, and control actions, which has driven the development of Intelligent Logistics. Although research has made strides in technological advancements and optimization algorithms, the development of comprehensive Performance Assessment frameworks—critical for evaluating and advancing these innovations—remains underexplored, often relying on limited or isolated metrics.This dissertation addresses this gap by assessing performance across three key dimensions: efficiency, user experience, and their interactions. Efficiency examines operational speed and resource utilization, while user experience highlights ease of use and service quality. Analyzing the dynamic interplay between these two helps find an optimal balance between system performance and user satisfaction. Our key insight is that integrating external factors, such as real-time traffic and user feedback, allows the system to consider environmental factors beyond its internal constraints. Specifically, we introduce three core systems:i) AdaTrans, focused on efficiency, incorporates external records from other companies as environment detectors to improve real-time transportation time prediction for individual orders.ii) COCO+, centered on experience, integrates historical and real-time behaviors of couriers and customers to improve causes identification in canceled pickup orders.iii) AdaService, emphasizing interaction, assesses last-mile terminal station performance, uncovering strategies to balance efficiency and service quality across varying operational conditions.These systems were tested with real-world data from a real-world logistics company. The results reveal that integrating external feedback and their interactions transforms the evaluation of logistics performance, surpassing state-of-the-art methods across diverse real-world scenarios.

: 

Contact  Professor Desheng Zhang

Zoom link: https://rutgers.zoom.us/j/5034279181?pwd=U2FFOGIwVkJJV1dHbzYxK2VTVDBiZz09