Taking vehicle behavior data to the next level: New vocation classifications unveiled
New vocation classification structure allows users to dive deeper into vehicle behavior.
With access to millions of data points and information at their fingertips, traffic planners and transportation departments are regularly diving into vehicle behavior, trying to make sense of the ‘why’ behind emerging trends and traffic patterns.
Analysts are also looking into why vehicles are on the roads, how they are operating and what purpose they are serving. It’s this kind of additional context that is becoming so crucial for transportation planners and engineers to make effective decisions and implement infrastructure changes that positively impact road networks and the citizens and fleet operators using them daily.
But while these are important questions to answer, it’s also important to not lump vehicles together for the purposes of analysis without careful consideration into those groupings. Instead it’s useful to dive deeper into traffic patterns to see where more natural classifications might arise, like with vehicle vocation.
Vehicle vocation explained
Vocation refers to a vehicle’s purpose on the road and it encompasses vehicle behavior as it goes about serving that purpose.
Originally developed to help analysts benchmark their vehicles in terms of safety or even sustainability metrics, vehicle vocation from Geotab ITS was a means to group like vehicles together in order to understand vehicle behavior in a like-to-like environment. Vehicle vocation measures how a vehicle is used independent of other analysis classifications or groupings like vehicle type, industry or ownership.
It was helpful to go beyond traditional group classifications like industry because vehicles in the food industry for example might behave differently depending on their vocation. Pizza delivery vehicles that operate in a hub-and-spoke pattern will obviously differ greatly from catering vans that have only one stop and might do lots of highway driving. This is where vocation classifications attempt to group vehicles together by behavior instead.
Vehicle behavior that speaks for itself
Initially, Geotab ITS relied heavily on customer input to inform the early vehicle vocation classifications used in the Altitude platform. Machine learning models were trained to look for similar patterns in vehicle behavior and resulted in unique vocations picking up on subtle patterns. For example, even the minor differentiations between hub-and-spoke long idling vs hub-and-spoke short idling vehicles were classified uniquely. Perhaps a sign of a public safety vehicle stationed in one spot for a night vs. a patrol vehicle circling their zone.
However, since the launch of Geotab ITS we now have better insights into which concepts of vocation are of use to our customers and so we set out to improve upon the classifications and models based on their feedback. Instead of attaching known vehicle use cases to fit predetermined vocations, we made the assumption that if distinct vocations exist, they would present themselves as clusters of similarly behaving vehicles naturally.
As a result, we now have 5 general use, clearly defined vocations.
Introducing the new vocation classifications for Geotab ITS
The classifications are extremely useful for drilling down into aggregate vehicle behavior for a deeper grasp of transportation patterns and behavior.
Here’s a couple examples of how these classifications can help add a new level of insight. For analysts interested in understanding how goods are moving from ports and across state lines, they can now use our trip chaining algorithm coupled with the ability to specifically look at long distance trucks to dive into potential supply chain issues and freight planning.
For city planners interested in a better understanding of how much last-mile delivery and on-demand services are impacting their downtown core, they filter specifically for those vocations and road segments of interest. Or consider the impact now due to the rise in e-commerce and how big box retail stores are no longer seeing the foot traffic they were used to and instead often serve as a depot to ship local goods to surrounding neighborhoods. This would result in a steady pattern of heavy-duty truck activity delivering the goods but also an increase in hub-and-spoke commercial traffic to and from the retail store.
The newly condensed vocation classifications reflect distinct clusters of similarly behaving vehicles to uncover aggregate traffic pattern insights. As we track more metrics, we may identify further clusters to extend the list.
Today, we have 5 vocation classifications for analysis in the Altitude platform:
These classifications are clearer and more relevant for running analysis and are more representative of aggregate vehicle behavior in order to base transportation planning decisions on.
We understand that this may impact current users of the Altitude platform who have run previous vocation analysis. For more details on the new and improved models and what this change means for your previous work, please contact email@example.com.
If you are not a current user and have questions about vehicle vocation data, please book a demo or reach out to us at firstname.lastname@example.org.