Module II – How do you measure fleet safety scores?
On the road to fleet safety, if you’re not keeping score, how can you build public trust?
In an era of heightened accountability and shrinking budgets, public fleet managers are under immense pressure to do more with less. Every decision is scrutinized and safety is paramount. But how can you demonstrate your commitment to safe fleet operations and build public trust without a clear way to measure and track your performance? The answer lies in a comprehensive fleet safety score.
Fleet safety score
What is a fleet safety score?
Think of a fleet safety score as a holistic report card for your fleet’s safety. It's compiled by gathering data from various sources on key elements of unsafe driving, each providing a piece of the puzzle. It provides a quantifiable indicator of your overall safety performance. This data-driven approach allows you to identify areas for improvement, track progress over time and showcase your dedication to protecting your drivers, your vehicles and the community you serve. In essence, a strong safety score is a testament to your responsible stewardship of public resources. Furthermore, it fosters transparency and accountability, demonstrating to the public that their safety and tax dollars are being managed with the utmost care.
How are fleet safety scores calculated?
Fleet safety scores are typically an amalgamation of individual driver safety metrics organized first at the driver level and then at the fleet level.
- Driver safety scorecards: the average has traditionally been used to calculate a fleet safety score.
- Predictive collision risk score: due to the limitations of driver scorecards along with advances in artificial intelligence (AI), we expect fleet safety scores will increasingly be measured using predictive collision risk scores.
- Collision detection: while not a true metric for calculating a fleet safety score, it can be used to assess a fleet’s real-time safety performance.
Driver scorecards
What are driver scorecards?
As discussed in Module I, fleet safety begins with understanding, selecting and tracking safety metrics. Driver scorecards use historical vehicle data to score drivers based on various driving behaviors, such as speeding, harsh braking and aggressive acceleration multiplied by the “weight” of that type of metrics.
Scorecards are a valuable tool for assessing and tracking the performance of individual drivers within a fleet. These scores can be used to identify areas where drivers need improvement and to track their progress over time. By comparing the performance of different drivers, fleet managers can identify top performers and those who may require additional coaching or training.
This data-driven approach can help improve driver safety, reduce collisions and optimize fleet efficiency.
Did you know?
The California Department of Conservation (CDOC) reduced safety violations from 30-40 per month to just 0-2 using Geotab’s telematics safety features to support the department’s safety program. CDOC drivers were provided access to their own dashboards and safety scorecards, giving employees direct feedback on their driving habits and safety adherence.
Building blocks of driver scorecards
The building blocks of driver scorecards are the various metrics used to assess and track driver performance. They are typically built leveraging a telematics platform, which collects and analyzes vehicle data to generate safety scores and identify areas for improvement.
Without a telematics solution, calculating a fleet safety score is a multi-step process that involves gathering vehicle data, normalizing the data to keep all metrics comparable, weighting the importance of different metrics and aggregating the results into a single, meaningful score.
The specific metrics used will vary depending on the priorities of the fleet, but the most common metrics for driver scorecards include:
- Speeding: Frequency and severity of exceeding speed limits.
- Harsh braking: Instances of sudden or forceful braking.
- Aggressive acceleration: Rapid acceleration that can lead to loss of control or increased fuel consumption.
- Harsh cornering: Taking turns at excessive speeds, increasing the risk of rollovers or loss of control.
- Seat belt usage: Whether or not the driver is wearing a seat belt.
To the right, you can see an example of Geotab’s Driver Safety Scorecard.
.svg)
Not all metrics are thought of equally
Different weights can be assigned to each metric in your safety score to reflect its relative importance to your fleet. Weightings are expressed as a percentage out of 100%. For example, if a fleet finds that there have been injuries or fatalities as a result of not wearing seatbelts, this would signal to the fleet manager that they should consider increasing the weighting of seat belt usage relative to the other metrics in their driver scorecards.
Limitations of driver scorecards
While driver scorecards offer valuable insights into past performance and many fleets have made meaningful safety improvements at the driver and fleet level, they do have some limitations. Driver scorecards only tell a limited story about past behavior. They don’t incorporate past behaviors such as impaired driving, lane departures or collisions. They also can’t tell what is happening on the road now or what is likely to happen.
Our research has found that, because of these limitations, driver scorecards are effective but tend to offer diminishing returns, as seen on the graph.
The limitations of driver scorecards include:
- Historical data: Scorecards primarily rely on historical data, and therefore only tell you part of the story. For instance, they will not reflect current road conditions.
- Limited scope: Scorecards do not capture all aspects of driver safety, such as distracted driving fatigue or road conditions.
- Lack of contextualized data: Every driver’s route, conditions and safety risks are different, but scorecards treat all drivers the same. For example, changing lanes in busy traffic is recognized as a risk factor for collisions, so a driver in Los Angeles, where drivers are frequently using freeways, has inherently higher risk than a driver on a country road. Scorecards can’t account for these differences.
- Lack of standardization across fleets: While the metrics leveraged for scorecards are in the most part standardized across fleets, the weights can be arbitrarily changed by fleets causing period-to-period comparison issues and the inability to accurately benchmark against similar fleets.
- Diminishing safety returns: Research findings from Geotab’s Data and Analytics team shows using driver scorecards can lead to lower collisions in the short term, but results often reach a plateau and then resist further improvements.
Despite these limitations, driver scorecards remain a valuable tool for improving fleet safety. By providing insights into driver behavior, they enable fleet managers to identify and address risky driving practices, ultimately leading to safer roads and reduced costs.
“You can’t drive forward by only looking in the rearview mirror.”
Or so the saying goes. To truly unlock a proactive safety program, fleets need to start looking for telematics partners that leverage AI to transform and analyze your data into predictive safety models, such as predictive collision risk scores.
Predictive collision rate
Fleet safety strategies have undergone a significant transformation in recent years. In the past, organizations primarily relied on historical data analysis to assess and improve safety. While providing valuable insights into past incidents and driver behavior, this "rearview mirror" approach has important limitations. It often results in reactive rather than proactive measures, addressing issues only after they occur, and offers limited preventative action.
Today, the landscape of fleet safety has evolved with the advent of telematics platforms that incorporate machine learning. These advanced systems enable proactive, near-immediate collision detection and notifications, enabling fleet managers to quickly respond to incidents and to identify high-risk drivers and situations.
Looking ahead, the future of fleet safety is increasingly focused on predictive analytics. This involves the integration of historical data with AI algorithms to forecast collision risks based on a variety of factors, including driver behavior, road conditions and environmental variables. This predictive approach empowers organizations to implement preventative measures, such as targeted training programs and optimized route planning, effectively shifting from a reactive to a preventative safety management paradigm. The shift from historical data analysis to near-immediate monitoring and predictive analytics signifies a fundamental change in how fleets approach safety, offering greater opportunities for risk mitigation and proactive management.
What is predictive collision rate?
Predictive collision rate is a forecast of the likelihood of a future collision per 100,000 units of distance (M/KM). This goes beyond simply analyzing past performance and instead uses current data to anticipate what might happen next. This could include predicting potential collisions, identifying high-risk drivers or forecasting maintenance needs. The ultimate goal being to provide a comprehensive understanding of your fleet's past performance and future risks will allow you to proactively mitigate potential issues and enhance overall safety.
What are the building blocks of predictive insights?
Predictive insights are built on a foundation of accurate and reliable data. Telematics devices and other sensors collect massive amounts of information on vehicle and driver behavior, such as speed, braking, location and engine diagnostics. This data is then processed and analyzed using machine learning algorithms and artificial intelligence to identify patterns and trends. The quality of the data and the sophistication of the algorithms directly impact the accuracy and reliability of the predictive insights.
An example of such technology is Geotab’s Predictive Risk model. The predictive insights go beyond traditional driver scoring models, proactively forecasting collision risk by analyzing a number of over 50 metrics such as:
- Previous driver behaviors
- Vehicle purposes and usage rates
- Types of vehicles and their weights
- Route disruptions, like construction work
- Weather and other environmental elements
- Time of day
.svg)
Did you know?
Geotab's predictive analytics provide a clear assessment of the likelihood of a fleet vehicle being involved in a collision within the next 100,000 miles or kilometers driven. Early results show that regular use of Geotab's Risk Analytics features can lead to a 5.5% reduction in predicted collisions. The Safety Center utilizes extensive real-world data models to provide actionable insights into collision risk, driver behavior and vehicle use, going beyond traditional benchmarking methods.
What are the benefits of predictive insights?
- Improved driver performance: Predictive insights can help identify drivers who are at a higher risk of being involved in a collision, allowing for targeted interventions and coaching to improve their driving habits
- Improved safety: By predicting the likelihood of future collisions, fleet managers can take proactive steps to mitigate risks, such as through targeted driver training, rather than be reactive.
- Enhanced decision-making: Predictive insights provide valuable data and insights that can support or inform decisions about safety programs, resource allocation and risk management strategies.
- Lower fleet costs: By reducing collisions and improving driver performance, predictive risk scores can contribute to lower operational costs.
- Improved fleet benchmark accuracy: Predictive insights tell you the probability of a collision over the next 100,000 miles or kilometers for your overall fleet and individual assets. This simple but powerful metric allows benchmarking against not only similar fleets, but also with similar assets and drivers in other fleets. For more details on fleet benchmarking, view Module III.
Overall, predictive insights empower fleet managers to move beyond reactive management and adopt a proactive approach that prioritizes safety, efficiency and cost-effectiveness.
Did you know?
Geotab customers using our AI-enabled and integrated safety features saw as much as a 40% reduction in collision rates. With our comprehensive driver safety apps, drivers can view their metrics and proactively participate in safety programs and hold themselves accountable.
Minor and major incident rates
While scorecards can be thought of as looking in the past and predictive insights as looking into the future, what about looking at the present? A vital safety metric to keep track of in real-time is both minor and major incident rates. Incident or collision detection technology has come a long way since the rise of Artificial Intelligence and machine learning. Geotab’s platform even works to reduce false positives and is now finding smaller collisions that might’ve previously gone unreported.
Did you know?
Advanced incident detection systems, such as Geotab’s, leverage the power of AI to distinguish between minor (1.5-2.5 Gs) or major (>2.5 Gs) incidents with great accuracy. With the capacity to determine the severity of incidents and near-immediate incident notifications, fleet managers can minimize the impact of collisions, gather the evidence they need in case of litigation and expedite the repair process.
On the road to safety, measuring your fleets’ risk assures your communities you are using a map
In the realm of public administration, fleet safety is not just a matter of compliance; it's a matter of public confidence. Despite its importance, public fleet managers face increasing pressure to find a route that balances safety with fiscal responsibility. Embracing a data-driven approach to inform a comprehensive fleet safety score can help you find the optimal route to effective and efficient safety operations.
AI is transforming fleet safety by providing predictive insights that enable proactive interventions and a shift toward preventative safety management. By incorporating predictive insights into their safety programs, public administration organizations can significantly improve driver safety, reduce collisions, mitigate liability, build public trust and optimize resource allocation.
For more information on the role of AI in safety, visit Module VII to watch Mastering the Art of Fleet Safety, a webinar on how AI-supported tools can analyze driver behavior, road risk and detected collisions to predict collision probability, identify high-risk areas and eventually supersede driver scorecards.
Article
Accelerating ROI Through Trusted Data Insights
In 2023, fleets enabled with Geotab’s AI-driven solutions and using Geotab’s integrated safety features saw a 40% reduction in collision rates, demonstrating a potential of 3,500 fewer collisions.