Session: 16-01-01: Government Agency Student Poster Competition

Paper Number: 149722

149722 - Managing Capacity for Curb Spaces in Urban Transportation System 

With the rise of new mobility operators and the increased demand for goods delivery services, especially post COVID-19, the need for efficient pickup/drop-off and loading/unloading access to curb space has grown significantly. This increase in demand has introduced complexities in managing curb space within urban transportation systems. Traditional uses of curb space, such as parking, now compete with these emerging needs, leading to increased congestion and inefficiencies. Proper management of curb space can reduce traffic congestion, enhance the efficiency of goods delivery services, and support the operations of new mobility providers. Thus, effectively allocating curb space is crucial for improving the overall traffic system.

Managing curb space presents several challenges due to the dynamic and diverse nature of urban environments. Balancing the competing demands for various uses requires incorporating different peak times and usage patterns. The variability in curb space capacity and demand necessitates management strategies capable of effectively allocating curb space in real-time. Research on curb space allocation for different uses is relatively new and still in its infancy. In practice, cities predominantly use curb spaces for parking, with allocation efforts for other uses often based on trial and error. This highlights the urgent need to develop effective policies for managing and allocating curb spaces. Current efforts and studies related to curb space management are primarily limited to data collection, focusing on understanding and observing the existing conditions. Pilot studies lack a decision framework and analysis of optimal policies. Moreover, existing data on curb space usage is limited and poorly organized. To address these challenges and fill the gap in the literature, this study aims to answer the following operational questions:

·       What are the numbers of vehicles served under steady-state conditions for each use?

·       What should be the optimal curb space allocation policy that maximizes the long-term running profit of the city?

To answer these questions, we analyze the interaction between curb space and traffic flow, which are critical elements of the transportation system that impact each other and affect the overall system performance. We first construct an open migration network to understand the flow of the vehicles by considering various uses of the curb space (i.e., parking, pickup/drop-off, and loading/unloading). We then build a newsvendor model based on the results derived from the migration network. Our goal is to maximize city profit by effectively distributing curb space among different uses.

Through numerical experiments, we study the curb space network using parameters estimated from the city of Detroit. Our developed optimal capacity allocation policies reveal that an increase in curb space capacity is necessary to maximize the total earnings, as the existing curb spaces fail to meet the optimal capacity allocation requirements. Subsequently, we simulate various scenarios with different model parameter settings and compare the results of our proposed policies with several common benchmarks. The results demonstrate that our developed policies allocate the curb spaces more effectively and practically. Additionally, the results highlight the importance of considering the proportion of vehicles searching for various curb spaces when formulating capacity allocation policies. Overall, the developed models and optimal policies enable more systematic capacity allocation decisions for various curb uses, resulting in significant improvements in the overall traffic system.

Presenting Author: Yuchi Guo University of Michigan - Dearborn

Presenting Author Biography: An IMSE Ph.D. candidate at UM - D, whose research interest is in operation research - supply chain management.

Authors:

Yuchi Guo University of Michigan - Dearborn
Armagan Bayram University of Michigan - Dearborn