AI applications using Open Road Crash data
Jurisdiction: Victoria
#Innovative solutions for Analysis, Risk Control and Presentation of Road Crash Statistics
How might we leverage road crash statistics and multi-agent AI-based web applications to enhance road safety and inform policy making?
Develop solutions to predict, prevent, and respond to road accidents. Use data from road crashes, traffic patterns, and environmental conditions to think creatively about how to use AI.
By integrating multi-agent AI systems, create a responsive solution to assist the understanding of:
- the causes of road accidents
- improving traffic management
- and formulating evidence-based policies.
The goal is to develop a solution that provides actionable insights to enhance road safety for all users, from pedestrians to drivers.
Develop, a multi-agent AI based web-application to provide the associated documentation, alongside notebooks for sample use-case walkthroughs.
These agents should be able to perform:
- Crash hotspot identification: An agent that can predict high-risk areas for crashes. And access to underlying data-driven models.
- Executive brief creation: An agent that creates comprehensive and concise executive briefs. Those briefs can summarise key findings, insights, and recommendations.
Recommended agent capabilities:
- Preventative measure suggestions: An agent capable of turning insights into effective measures. Examples include changes in road design or traffic regulations to reduce crashes.
- Policy guideline recommendations: An agent that can suggest creative public policy solutions. Examples include:
- improved road signage
- better lighting
- enhanced traffic signal timing
- public awareness campaigns
- or infrastructure modifications.
- Historic data insights presentation: An agent capable of analysing historical data. to understand crash patterns and suggest improvements.
Submission Requirements
- GitHub Repository (or similar): Share a link to a GitHub repository containing:
- all project files
- documentation
- code
- and system design diagrams.
- Executive Summary: A concise summary of the project, key findings, and recommendations.
- Model Binaries (if applicable): Submit the model binaries with the GitHub repo if applicable.
The Victorian Road Safety Strategy 2021-2030 seeks to embed a strong culture of road safety within the community, supporting road users to make safe choices as we progressively reduce fatalities on our roads.
Image Credit: © Copyright State of Victoria (Department of Transport and Planning), Creative Commons Attribution 4.0 International licence.
Compulsory datasets
Crash statistics data
https://discover.data.vic.gov.au/dataset/victoria-road-crash-data
Executive Brief Format
https://discover.data.vic.gov.au/dataset/0b428714-ef9b-4ac7-a8c5-815685c40dff/resource/e35b5d5a-8a98-4a82-8bdc-c9c93f0ac6aa/download/brief-template-for-gov-hack-2024.pdf
The above datasets are also provided below as links along with other recommended datasets.
Eligibility: Open to everyone. See the expected outcomes and the use of Crash Stats data listed above
Entry: Challenge entry is available to all teams in Australia.
Dataset Highlight
Traffic Count Locations - Victoria Roads
Traffic Lights
Traffic Signal Configuration Data Sheets
Traffic Signal Volume Data
VicRoads Turning Movement Volume Surveys
Traffic Volume
Speed Zones
Lane Use Management Signals
Multiple Variable Electronic Speed Signs Live
Speed Signs
Vicmap Datasets
GovHack 2024 Brief Template
Victoria Road Crash Data