Keeping our roads safe

A Framework for Analyzing Road Accidents In Thailand Using Machine Learning Models

Comment

Inspiration

What it does

Globally, Every year the lives of approximately 1.3 million people are cut short as a result of a road traffic crash (WHO,2021).

The number of casualties, injuries, and deaths caused by traffic accidents each year and the significant economic costs involved make traffic accidents one of the most prominent global concerns.

Road accidents are caused by a variety of factors. If these factors can be better understood and predicted, it might be possible to take measures to mitigate the damages and their severity.

How we built it

Challenges we ran into

This is my first time to build machine learning model using Azure. It's quite a steep learning curve to implement and publish the model. In the beginning, I encountered frequent issue related to the azure account. Similarly, I dont have access to power BI due to device issue, hence I have to borrow my auntie laptop to run the analysis and visualisation on power BI.

Accomplishments that we're proud of

I'm glad and grateful for this learning opportunity and utilise my marketing insights knowledge in exploring & triangulate all possible datapoints to analyse trends of accidents.

What we learned

This case study calls for further steps in which dynamic hotspots can be identified so that this information can be used for better management of public and private resources. The effective identification of dynamic hotspots can be used in a number of use cases including allowing:

  • local municipal / city authorities to more effectively organise their traffic control systems
  • Insurance companies to share this information with their clients for a safer driving experience
  • end-user app such as Waze to optimize routes by anticipating dangerous spots

What's next for Keeping our roads safe

This current study faced several constraints, such as limited to only four databases from Injury surveillance system owned by Division of Injury Prevention, Department of Disease Control, Ministry of Public Health, Thailand. I'm also planning to predict traffic accident hotspots with spatial data

The datasets focus on the injured people that have got treatment in hospitals; and disregard the fatal accidents data. For further research can be supplemented by utilising road type data for model training. This will improve the prediction of traffic accident and further enhance the accuracy of predictive models, addressing the responsible factors for accidents likely to happen.

Built With

Share this project:

Updates