已编辑 作者：EishiFUN
I'm interested in how other fleet managers approach rules for events such as harsh braking, speeding or excessive idling. Do you rely mainly on the default thresholds, or have you found that adjusting the conditions and duration makes the alerts significantly more useful?
I'd also be interested to hear how you balance getting enough alerts to identify problems without overwhelming drivers or fleet managers with notifications.
Thank you
Alain Fleurs Vevey
CCO
You need to be logged in to reply to this post and participate in the Geotab Community discussions.
Hey @patrickbarneaux
Thank you for asking this question. Glad you came to our community.
I would approach this in three parts:
Speeding and idling
Start with the fleet policy you want to enforce, then use a duration to avoid reacting to very brief events. For example, a speeding rule might require the vehicle to exceed the limit by a defined amount for a minimum number of seconds. Idling can be handled similarly.
It can also help to create graduated rules, such as 5, 10, and 20 minutes of idling, but only send notifications for the higher-severity events. The lower-severity rules can remain available for reporting and trend analysis.
Harsh braking and acceleration
These rules are more dependent on vehicle type, road conditions, and driving context. The older sensor-based rules can produce more false positives, especially when applied to a mixed fleet.
Where available, I recommend starting with the newer GPS-based harsh event rules. These use different thresholds based on vehicle class and are designed to reduce noise from events such as road conditions. If you still see unexpected results, review the vehicle classification, GPS data quality, and the surrounding driving context. Camera footage can also be useful when you need to understand why a specific event occurred.
Notification strategy
It is often better to separate rule detection from notification. First configure the rule and review the resulting events using alerts such as Log only or through reporting. Once you understand the event volume and confirm that the rule is useful, add email, popup, or driver notifications for the events that require action.
The Collision Risk reports can also provide useful context by benchmarking driving risk against similar vehicles and operating conditions. This is complementary to exception rules, since it helps identify broader patterns rather than only individual events.
The goal is not necessarily to eliminate all exceptions. It is to make sure the exceptions generated are meaningful, aligned with your safety policy, and actionable for the people receiving them.
I hope that helps!
Eishi FUN
Hi
Regarding the rules you mention, yes, we depend on the behavior of the accelerometer, but I suggest that when setting up the rules, you take into consideration the thresholds they offer you for each type of vehicle.
Hey @patrickbarneaux
Thank you for asking this question. Glad you came to our community.
I would approach this in three parts:
Speeding and idling
Start with the fleet policy you want to enforce, then use a duration to avoid reacting to very brief events. For example, a speeding rule might require the vehicle to exceed the limit by a defined amount for a minimum number of seconds. Idling can be handled similarly.
It can also help to create graduated rules, such as 5, 10, and 20 minutes of idling, but only send notifications for the higher-severity events. The lower-severity rules can remain available for reporting and trend analysis.
Harsh braking and acceleration
These rules are more dependent on vehicle type, road conditions, and driving context. The older sensor-based rules can produce more false positives, especially when applied to a mixed fleet.
Where available, I recommend starting with the newer GPS-based harsh event rules. These use different thresholds based on vehicle class and are designed to reduce noise from events such as road conditions. If you still see unexpected results, review the vehicle classification, GPS data quality, and the surrounding driving context. Camera footage can also be useful when you need to understand why a specific event occurred.
Notification strategy
It is often better to separate rule detection from notification. First configure the rule and review the resulting events using alerts such as Log only or through reporting. Once you understand the event volume and confirm that the rule is useful, add email, popup, or driver notifications for the events that require action.
The Collision Risk reports can also provide useful context by benchmarking driving risk against similar vehicles and operating conditions. This is complementary to exception rules, since it helps identify broader patterns rather than only individual events.
The goal is not necessarily to eliminate all exceptions. It is to make sure the exceptions generated are meaningful, aligned with your safety policy, and actionable for the people receiving them.
I hope that helps!
Eishi FUN