Road Monitoring

Road safety is a critical issue that affects millions of people every day. With the increasing number of vehicles on the road, incidents and accidents are becoming more common. These lead to serious injuries, loss of life, and economic damages. To address this issue, the Dutch government partnered with car-manufacturers on an enormous data project. This Road Monitoring project – or RoMo in short – is a data-driven initiative to improve road safety and traffic flow.

This project was the first push by the government for their broader RoMo project. In this initial stage, they needed insight into what their data could do for them, whether they should continue with their project, and how. AI Heroes was brought on board to help the government understand the value of their data and determine the next steps in their data strategy.

Why Care About Data?
Data is the foundation of many AI applications, such as chatbots, computer vision, natural language processing, and machine learning. Data analysis allows for insights that are not visible to humans, by making countless comparisons and calculations. Using the findings obtained from data, it leads to informed decision making, improving business intelligence. This can effectively reduce cost, resources, and lead to more satisfied customers as well. Additionally, data analysis can be used to predict and forecast scenarios, behavior and trends. Once a data analysis framework is established, it’s easy to keep using it for updated analyses and continuously improve performance.

Uncovering such value was a main ambition for the Ministry of Infrastructure and Water Management from their new dataset. To achieve this, we explored two datasets and combined them for maximal predictive power. Their own dataset contained reported incidents, such as accidents, objects on the road, or broken down vehicles. The other dataset contained warning signals coming from cars on the road, such as warnings for slippery roads, emergency braking or car crashes.

The Power of Data for Road Safety
For the Ministry of Infrastructure and Water the value of their data lies in the ability to know what causes accidents and when they are most likely to happen. This helps them to improve public safety and their own efficiency by leveraging the power of data and AI.

• With a hotspot analysis they get insights into the types of accidents that happen, for what reason they happen, and they can act on it! Numerico’s data experts are able to display where on the map incidents are most likely to happen and under what conditions.
• With the predictive power of AI they know better when and where to deploy resources, such as people and vehicles. This will allow them to prepare, shorten response times and save costs as they deploy the right amount of resources at the right place.
  With our data science project, the government can make better decisions, save lives and resources, and enhance their public service.

How do we find the value in the data?
This new dataset contains information about the safety warnings that are collected by cars, such as collisions, strong crosswinds and slippery road conditions. To find the value of this additional dataset we had to combine it with their existing dataset on actual incidents and accidents.

To achieve this, we worked together with our partner Numerico to clean, label and preprocess the data. As this project was all about creating understanding of what the data was and could do we visualized it for their decision-makers with graphs, dashboards and maps. This was achieved by using language and visuals that were easy to understand for non-data scientists and data scientists alike. To give a few examples, graphs were used to show during which times unsafe road conditions were most likely to occur and which risks were present. Maps were a very beneficial tool as it could be used to pinpoint those roads, intersections and which turns cause the greatest risks under certain conditions.

On to the Next Step
The task of this project was to identify if their new datasets held value for data analysis and if they should continue with their data collection efforts. As the preliminary results suggest that the new data can be used to predict and better understand incidents happening, the answer was that their project was indeed very valuable and should be continued. This project will continue in the future, but our current results provide statistically underlied insights, and a system to continue the project immediately when more data is available. With the help of our data-based approach, we were able to provide insights that would be impossible to achieve otherwise. With a bigger, follow-up project it seems likely that in-car data can be used to predict and even avoid road incidents. Continuing this project can result in fewer incidents, as well as seriously speeding up incident management and knowing where and when to commit their valuable resources.

How can AI Heroes help you?

• Develop a data strategy: If you want to revolutionize the way your organization uses data, we can help you unlock the value of your data with a data audit and build a strategy on how you can best use your data to your advantage.
• Help you make sense of your data: Many clients have a lot of data, but want to find out what that data can mean for them. AI Heroes can help you explore your data, find patterns and insights, and help you understand your data using visualizations and reports.
• Monitor and improve AI performance: AI Heroes can help you measure the impact and accuracy of your AI solutions, and provide ongoing support and maintenance to ensure they stay relevant and reliable.