时间：2024年5月13日（周一）下午14:00~16:00

地点：通达馆103会议室

讲座一主题：How to make micro-delivery services more efficient? Perks and pitfalls of combinatorial optimization and AI-based methods.

主讲人：Shadi Sharif Azadeh

邀请人：安琨 教授

主讲人简介：

Shadi Sharif Azadeh is an associate professor at Civil Engineering and Geosciences faculty and the co-director of SUM (Sustainable Urban Multi-modal Mobility) lab at TU Delft in the department of Transport & Planning. She holds a PhD in Mathematics (operations research) from Polytechnique Montreal where she received doctorate excellency award at University of Montreal (CIRRELT) as well as Michael Florian Award for best PhD thesis research award in Canada.

Her areas of expertise include integration of operations research with behavioural models for transport, mobility and logistics networks (Choice Driven Optimization). Her current major projects are related to i) developing methods to tackle uncertainty with a special focus on forecasting and scenario generation for passenger mobility and parcel delivery, ii) combining pricing and assortment optimisation methods to model supply and demand interplay for last mile delivery and urban mobility systems iii) developing real-time methods to incorporate in combinatorial optimisation models for large-scale transport problems iv) special focus on designing sustainable multi-modal transport systems v) introducing solutions to mitigate the impacts of climate change on transport networks.

She is an associate editor of Transportation Science journal, editorial board editor at Transportation Research Part B: Methodological, editorial board member of Transportation Research Part C: Emerging Technologies and associate editor of Nature Series (npj) Sustainable Mobility and Transport journal. She has served as guest editor of three special issues at Transportation Science (2021-2023), EURO Journal of Transport and Logistics (2021-203) and OR Spectrum (2023).

Currently, she is the PI of SINERGI (Sustainable Innovative digitalized NEtwork of uRban loGIstics) project funded by JPI-ERANET focused on micro-delivery for city logistics. She is acting coordination team member of the Horizon Europe project called Seamless Shared Urban Mobility (SUM) that aims at transforming the current mobility networks towards innovative and novel shared mobility systems (NSM) integrated with public transport (PT) in more than 15 European Cities by 2026 reaching 30 by 2030.  She is receipent of a direct funding from Rijkswaterstaat dedicated to mitigate the impacts of climate change on transport network.  She serves in the coordination team of the NWO Perspectief funded project called  XCARCITY that aims to develop realistic digital replicas of car-low areas in Amsterdam, Almere and Rotterdam.  She holds the same role in metaCCAZE (HE project) and V2G QUESTS (Driving Urban Transitions) projects.

主讲内容简介：

Microdelivery services are promising solutions for on-demand city logistics. To promote these sustainable tools of instant delivery services, their delivery planning needs to be more efficient. To improve delivery efficiency, on-demand meal delivery platforms seek to optimize real-time management of their courier resources based on anticipatory insights into demand distributions within the city. Accurate and realtime demand models are essential to these systems’ efficiency. Display optimization, downranking of restaurants in the shortage of couriers, behavioral models for both riders and consumers, routing suggestions while keeping the bike lanes safe are few of the challenges these services need to cope with. In this presentation, I will explain the pros and cons of combinatorial optimization methods and AI-based approaches to tackle these complex problems for the context of meal-delivery.

讲座二主题：Impact of Context-Aware Assortment Optimization in Platform-Based Urban Mobility

主讲人：Yousef Maknoon

邀请人：安琨 教授

主讲人简介：

Dr. Yousef Maknoon is an associate professor at the Faculty of Technology, Policy, and Management (TPM) at TU Delft. He is also the director of Orbit Lab, a research group specializing in Operations Research and Behavioral Informatics in Transportation. His research takes a multidisciplinary approach, rooted in operations research, to tackle emerging challenges in the transport and logistics domain. In recent years, his primary focus has been on the design and operational strategies for on-demand and instant logistics services.

主讲内容简介：

This study investigates the impact of context effects on assortment optimization in platform-based urban mobility services. We critique the classical assumptions inherent in Random Utility Maximization (RUM) models by employing an alternative framework, the Random Regret Minimization (RRM) model. Specifically, we introduce a novel strategy, termed the `Marginal Decoy Policy,' designed to address behavioral anomalies that classical models fail to capture. A comparative analysis is conducted with established models such as the Multinomial Logit (MNL) and Generalized Random Regret Minimization (G-RRM). We present evidence illustrating significant enhancements in customer satisfaction and retention by incorporating context effects.

欢迎各位老师、同学参加院级高等讲堂！

交通运输工程学院研究生会
交通运输工程学院青年教师沙龙