Improving road safety for heavy vehicle operations
“Databricks has been a game-changer for our data infrastructure. Our data science environment, using machine learning and real-time processing, runs on Databricks Data Intelligence Platform, minimizing the effort required and saving more time. The speed has allowed us to build quality products even faster.”
The National Heavy Vehicle Regulator (NHVR) is the statutory authority that oversees heavy vehicles in Australia. The NHVR works collaboratively with governments and its customers to drive safety, productivity and efficiency outcomes across the heavy vehicle transport sector and the Australian economy. Having accurate and real-time insights into the country’s fleet of nearly 1 million heavy vehicles is fundamental to identifying and regulating unsafe driving practices. Limited by the capabilities of its traditional data warehousing solution, the NHVR chose to elevate its digital capabilities with Databricks Data Intelligence Platform to analyze large volumes of data, including more than 4.5 million monthly vehicle sightings from around the country. Now it is able to deliver real-time (subsecond) reporting, dashboards and alerts, helping its safety and compliance teams to detect risks as they develop, and to carry out immediate interventions that could disrupt imminent danger to public safety.
Mitigating road safety risks with data and machine learning
The heavy vehicle industry is critical to the economy and continues to grow rapidly, creating greater potential safety risks for road users and the heavy vehicle workforce. The NHVR is a risk-based, intelligence-led regulator that aims to proactively target risks in the heavy vehicle sector to prevent fatalities and serious injury, rather than react to these events after the fact. Taking an evidence-based approach to regulation, the NHVR acquires heavy vehicle data to identify patterns that help predict risks and administer timely and effective intervention. Overseeing a national fleet of almost 1 million heavy commercial vehicles, the NHVR has been collecting high volumes of data to assess safety risks across the fleet. Vehicle sightings alone generate more than 4.5 million data points a month.
Real-time and high-speed data processing generating richer insights
For the NHVR, a data analytics platform that could handle large data sets and support its ability to scale machine learning and analytics was critical. But, beyond that, it needed a platform that supported real-time processing cost-effectively, to enable predictive analysis and faster time to insight, in order to disrupt in-flight safety threats.
“Like all government agencies, we are constrained by the resources we have. Sophisticated analysis helps us to optimize our resources to focus on vehicles, operators and entities that pose the highest risk to human life. It also helps us assess which interventions, ranging from education to prosecution, are the most effective in improving road safety,” said Tammy Wigg, Director of Data Science and Analytics at National Heavy Vehicle Regulator.
NHVR had developed a data acquisition strategy to better understand the industry it regulates, and, as a result, started to collect large amounts of data sets that its warehouse could not ingest quickly enough. As these large sets of data came in, it became critical for NHVR to have a platform that was able to handle and process them in a quick and seamless manner.
To cope with the high volume of new and unstructured data sets, the NHVR turned to Databricks Data Intelligence Platform, which also allows for its data science team to run machine learning experiments, quickly and easily. With the ability to create machine learning templates, less time was needed to implement those experiments, and the NHVR was able to process more work and gain valuable insights more quickly.
Deploying more targeted safety interventions at greater speed
With Databricks Data Intelligence Platform, the NHVR is now able to pull all data and analytics needs into a single, unified platform. It can now analyze large amounts of data in a shorter period of time, and gain predictive insights into the heavy vehicle industry in Australia to create more targeted safety interventions. Databricks also allowed the NHVR to detect risks in real time, enabling it to send alerts to safety and compliance officers in the field, to intercept vehicles potentially posing imminent danger to public safety.
Useful insights into vehicle mass, vehicle travel times, driver rest breaks and their duration, as well as other information, all contribute to the NHVR’s ability to more effectively regulate the Australian heavy vehicle industry and provide targeted education to its participants.
Out of this initiative, the NHVR has built a fatigue engine to help identify and target drivers and operators most at risk of causing fatigue-related accidents. This entails being able to process high volumes of data quickly to enable the regulator to direct appropriate interventions on the ground, and has resulted in a 4% increase in yielding high-risk fatigue infringements. By continuing to refine the fatigue engine over time and scale, the NHVR will be able to successfully target more high-risk drivers, resulting in safer outcomes for all road users. The NHVR has also built data engines that enable improved targeting of other safety risks, such as vehicle roadworthiness and over-mass vehicles.
In addition, the NHVR has created a more reliable crash prediction model powered by machine learning. By leveraging multiple data sets, including vehicle data, defect data and crash data, the NHVR was able to identify a cohort of vehicles and operators that had a 1-in-38 chance of being involved in a fatal or serious incident on any given day. This prediction is actionable, with the NHVR able to deploy its safety and compliance officers to target these specific vehicles on the road, conduct site visits and implement a range of other interventions to prevent serious incidents.
Databricks was integrated with NHVR’s existing Azure environment, optimizing the organization’s previous investment, and creating a more seamless data analytics platform. This meant that NHVR was able to further save costs and focus on leveraging data in a rapid manner for real-time and predictive insights.
“The Databricks Data Intelligence Platform’s accessibility, and the speed at which it enables us to obtain data and insights about safety risks, have enabled us to enrich our understanding of our industry and to provide more targeted and appropriate education and regulation. Every day, we strive to make Australia’s roads safer using the power of data,” said Wigg.
