Adoption of AI Analysis on Traffic Data for Traffic Prediction and Management

I&T Wish Adoption of AI Analysis on Traffic Data for Traffic Prediction and Management
(REF: W-0456)
Matched I&T Solution
Trial Project
Summary and Challenges
  • Traffic congestion has become a serious problem in Hong Kong, which affect the road safety and experience of road users. Volume prediction and detector failure is rarely been considered. The develop of deep learning AI traffic model to predict congestion level and traffic volume could be able to distribute the traffic to other route.
Expected Outcome
  • We propose to develop deep learning AI traffic model that can accurately predict traffic congestion levels in Hong Kong for a short time horizon of up to 30 minutes.
  • The model should be able to predict congestion levels on typical and atypical days, including major events and incidents.
  • Additionally, the model should be able to identify recurrent and non-recurrent congestion and establish unique congestion judgment criteria for each location and time slot.
  • The model should categorise congestion levels into three categories and the proposal should define the metric for each congestion level for different road types.
Expected Trial Duration 2-month
Contact Information
I&T Wish Proposer:Electrical and Mechanical Services Department (EMSD)
Contact Person:Cheng Chun Kit
Position:Project Officer/Innovation/34
Tel:3908 2572
Email: chengck@emsd.gov.hk
Initiating Department Transport Department (TD)
Upload Date 2023-05-17
Closing Date 2023-05-31
Propose I&T Solution Closed