This article details an interesting view of “Macroeconomic” level understanding of fleets using GPS data which we get here at GPS Insight. On a typical day for one set of our servers, the number of GPS data points from vehicles under management reaches around 700. This is indicative of the number of moving vehicles, and with two minute updates for these vehicles, it roughly means 1500 vehicles are moving (and we are tracking) at that time for those servers.

Here is a graph of the activity throughout a typical work week day, with a yellow “15 minute moving average.” The Pink lines indicate our users’ activity on the system, and the blue/yellow lines ultimately indicate their vehicle movement and/or idling time:


On a WEEKEND, however, the trend is much less smooth, and some interesting “spikes” occur which could show why you probably want GPS tracking for your vehicles if your drivers take them home on the weekend (we also see that the level our customers use GPS Insight is down considerably on the weekends):


You may want to click on the image above to show a larger version. The day appears to spike at only 600 or so vehicles moving (vs. 1500) at once, and while the day begins about the same time as a typical weekday, there are a couple late-day spikes.

I imagine these spikes illustrate just how many people use their company vehicles to drive to restaurants/bars, particularly the late night return spike which we never see during the week but which is prevalent on the weekends.

Running a quick back-end report, I identified some likely culprits and drilled down on one (there were 150 vehicles driving during that time frame). This driver, a salesperson by the vehicle’s label, went on a 11:45 diaper run from what it looks like, thankfully less than a mile from home, stopping 7 minutes, then heading back. We see that this is a “Fry’s Food Stores” by turning on the shopping overlay/layer which gives useful information about an area within the 3-D mapping we utilize for GPS Insight.

This type of company vehicle utilization may be completely fine with the customer, but it could also be something more dangerous (late night drinking and driving). To see if this vehicle engages in late-night usage we can run a simply “odd-hours” report:

And luckily, we see that in the last 30 days, the only activity this vehicle has had between 10PM and 2AM is this event to the store and a 1 minute movement, most likely moving the vehicle in or out of the garage.

This is a simple example on the first vehicle (of 150) I checked, and for customer privacy we won’t give out much information about others, but the spikes and the trends are compelling evidence that your drivers may be taking your vehicles out at night for fun on weekends, and with GPS Insight, you can easily detect and deter this type of behavior. After all, it’s your company on the line from a risk standpoint if an accident occurs, and minimally, the fuel and maintenance for off-hours/weekend usage is something our customers wish to avoid.

For more information about these types of reports and maps, visit GPS Insight or our Wiki with more examples and documentation about the GPS Insight Vehicle tracking product.

Rob Donat.