A simulation platform for shared mobility services

Principal Investigator: Dr. Jintao KE (Assistant Professor from Department of Civil Engineering)

This project is showcased in the second exhibition – Digitization in Innovation Wing Two

About the scholar

Dr. Jintao KE

Research interests:
• Smart mobility, smart transportation, and smart city
• Urban computing and transportation big data analytics
• Artificial intelligent for transportation
• Operations research and operational management for on-demand mobility
Email: kejintao@hku.hk
Website: https://sites.google.com/view/kejintao

Project information
This is a large-scale simulation platform for managing and controlling Hong Kong taxis. The simulator platform can be used to simulate the movements and trajectories of taxis for idle cruising, picking up passengers, and delivering passengers on a large-scale transportation network. The simulation platform is calibrated by a real dataset of Hong Kong taxis to ensure that the simulation well approximates the reality. This simulator is jointly developed by the teams of Dr. Jintao Ke at HKU and Prof. Hai Yang at HKUST. The simulation platform will be open for public use in the near future.

How it can be used?
Researchers can make use of this simulation platform to train and test optimization, machine learning and reinforcement learning algorithms for designing better operating strategies, such as order dispatching, vehicle repositions and dynamic pricing. The simulation platform can also assist the transport department of Hong Kong government in evaluating and designing policies for Hong Kong taxi markets, including vehicle fleet size control and pricing regulations. The ultimate goal of developing this simulation platform is to help taxi operators and government to better manage and operate taxis and ride-hailing vehicles (such as Uber) so as to achieve a sustainable, efficient, and environmentally friendly on-demand transportation systems in Hong Kong.

Project video
Project images
Dashboard view consists of 9 parts, the above graph is the preliminary display.Dashboard view

Finished orders during different period: This part shows the number of finished orders and cancelled orders during different period.

Evaluation metrics: This part shows the different metrics of our simulator, they are matching rate, pickup ratio, delivery ratio and idle ratio respectively.

Transportation: This part we plan to show the carrying passenger of different kinds of transportation.

Order finished status: This part will show the order status per day.

Different line color: This part shows the meaning of different line in the middle map animation.

Bridge flow rank: This part we plan to show the bridge flow rank of Manhattan.

Six metrics explanation

  • Passenger pickup time is the average pickup time of all orders within the current 30 minutes.
  • Passenger waiting time is the average waiting time of all passengers within the current 30 minutes.
  • Matching rate is equal to the current completed orders divided by generated orders.
  • Pickup ratio is equal to the current drivers which are picking up divided by total drivers.
  • Delivery ratio is equal to the current drivers which are delivering divided by total drivers.
  • Idle ratio is equal to the current drivers which are cruising divided by total drivers.
Line trip map
Node trip map
Car map
Achievement of the Project
Enquiry / Feedback

Please feel free to give your enquiry / feedbacks to the research team by filling the form (https://forms.gle/JV59N47nTj19ndYz6). Thank you!