Driver log retention system
The described features of the present disclosure generally relate to one or more improved systems for analyzing the electronic information associated with driving activities (e.g., driver log information) obtained from the one or more mobile computing platforms (ELDs) associated with one or more vehicles to identify a likelihood of a driver resigning or deserting his or her position. Accordingly, features of the present disclosure may identify “at-risk” drivers for the fleet operators to trigger remedial measures to prevent such adverse event (e.g., driver quitting).
1. A driver log-based retention system, comprising:
a processing system configured to:
receive, from electronic logging devices (ELDs), driver log information associated with drivers, wherein each ELD is configured to track the driver log information for Hours of Service (HOS) compliance with each ELD configured to receive a current driving state manually entered by a driver, the driver log information comprising information associated with a driver associated with one or more vehicles and includes shift start variance information;
aggregate the driver log information for a predetermined time period;
transform the aggregated driver log information into a set of derivations derived from the aggregated driver log information for each of the one or more drivers in accordance with one or more derivation rules associated with a turnover prediction model; and
apply the turnover prediction model to the set of derivations to generate a prediction report that identifies a retention condition of employment for one or more drivers who are at risk of resigning;
a transmitter configured to transmit a notification message, including the prediction report and an identification of the retention condition, to a scheduling engine; and
the scheduling engine automatically modifying assigned routes of at-risk drivers based on the prediction report and the retention condition that identifies the risk of one or more drivers resigning.
2. The system of claim 1 , wherein applying the turnover prediction model to the derivations to generate the prediction report comprises analyzing one or more predictor variables from the set of derivations, wherein the one or more predictor variables are selected from a group consisting of:
a total driving hours;
a shift start variance;
a number of early shift starts;
a number of late shift starts;
a number of HOS violations;
a number of on-duty not driving hours; and
a number of off-duty hours.
3. The system of claim 2 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
assigning a variable weight to each of the one or more predictor variables for use in an equation to identify the retention condition of employment, wherein the retention condition of employment identifies at-risk drivers.
4. The system of claim 1 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
determining a total driving hours during the predetermined time period for the one or more drivers based on the driver log information associated with each of the one or more drivers;
determining whether the total driving hours exceeds a threshold; and
generating the prediction report that identifies the retention condition based on the determining that the total driving hours exceeds a threshold for at-risk drivers.
5. The system of claim 1 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
identifying an average shift start time for the one or more drivers;
determining a shift start variance for the predetermined time period for each of the one or more drivers based on identification of the average shift start time;
determining that the shift start variance exceeds a shift variance threshold; and
generating the prediction report that identifies the retention condition based on the determining that the shift start variance exceeds a shift variance threshold for at-risk drivers.
6. The system of claim 1 , wherein extracting the set of derivations comprises extracting a subset of at least one of the driver log information that indicate that a respective driver is in danger of resigning.
7. The system of claim 1 , further comprising triggering a remedial measure for at least one of the one or more drivers based on a value of a confidence factor.
8. An apparatus for identifying at-risk drivers, comprising:
a processor; and
a memory coupled with the processor, wherein the memory includes instructions to:
receive, from electronic logging devices (ELDs), driver log information associated with drivers, wherein each ELD is configured to track the driver log information for Hours of Service (HOS) compliance with each ELD configured to receive a current driving state manually entered by a driver, the driver log information comprising information associated with a driver associated with one or more vehicles and includes shift start variance information;
aggregate the driver log information for a predetermined time period;
transform the aggregated driver log information into a set of derivations derived from the aggregated driver log information for each of the one or more drivers in accordance with one or more derivation rules associated with a turnover prediction model;
apply the turnover prediction model to the set of derivations to generate a prediction report that identifies a retention condition of employment for one or more drivers who are at risk of resigning; and
transmit a notification message, including the prediction report and an identification of the retention condition, to a scheduling engine that automatically modifies assigned routes of at-risk drivers based on the prediction report and the retention condition that identifies the risk of one or more drivers resigning.
9. The apparatus of claim 8 , wherein the instructions to apply the turnover prediction model to the derivations to generate the prediction report are further executable by the processor to analyze one or more predictor variables from the set of derivations, wherein the one or more predictor variables are selected from a group consisting of:
a total driving hours;
a shift start variance;
a number of early shift starts;
a number of late shift starts;
a number of HOS violations;
a number of on-duty not driving hours; and
a number of off-duty hours.
10. The apparatus of claim 9 , wherein the instructions to apply the turnover prediction model to the set of derivations to generate the prediction report are further executable by the processor to:
assign a variable weight to each of the one or more predictor variables for use in an equation to identify the retention condition of employment.
11. A method for identifying at-risk drivers, comprising:
receiving, from electronic logging devices (ELDs), driver log information associated with drivers, wherein each ELD is configured to track the driver log information for Hours of Service (HOS) compliance with each ELD configured to receive a current driving state manually entered by a driver, the driver log information comprising information associated with a driver associated with one or more vehicles and includes shift start variance information;
aggregating the driver log information for a predetermined time period;
transforming the aggregated driver log information into a set of derivations derived from the aggregated driver log information for each of the one or more drivers in accordance with one or more derivation rules associated with a turnover prediction model;
applying the turnover prediction model to the set of derivations to generate a prediction report that identifies a retention condition of employment for one or more drivers who are at risk of resigning; and
transmitting a notification message, including the prediction report and an identification of the retention condition, to a scheduling engine;
automatically modifying, by the scheduling engine, assigned routes of at-risk drivers based on the prediction report and the retention condition that identifies the risk of one or more drivers resigning.
12. The method of claim 11 , wherein applying the turnover prediction model to the derivations to generate the prediction report comprises analyzing one or more predictor variables from the set of derivations, wherein the one or more predictor variables are selected from a group consisting of:
a total driving hours;
a shift start variance;
a number of early shift starts;
a number of late shift starts;
a number of HOS violations;
a number of on-duty not driving hours; and
a number of off-duty hours.
13. The method of claim 12 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
assigning a variable weight to each of the one or more predictor variables for use in an equation to identify the retention condition of employment.
14. The method of claim 11 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
determining a total driving hours during the predetermined time period for the one or more drivers based on the driver log information associated with each of the one or more drivers;
determining whether the total driving hours exceeds a threshold; and
generating the prediction report that identifies the retention condition based on the determining that the total driving hours exceeds a threshold for at-risk drivers.
15. The method of claim 11 , wherein applying the turnover prediction model to the set of derivations to generate the prediction report comprises:
identifying an average shift start time for the one or more drivers;
determining a shift start variance for the predetermined time period for each of the one or more drivers based on identification of the average shift start time;
determining that the shift start variance exceeds a shift variance threshold; and
generating the prediction report that identifies the retention condition based on the determining that the shift start variance exceeds a shift variance threshold for the at-risk drivers.
16. The method of claim 11 , wherein extracting the set of derivations comprises extracting a subset of at least one of the driver log information that indicate that a respective driver is in danger of resigning.
17. The method of claim 11 , further comprising triggering a remedial measure for at least one of the one or more drivers based on a value of a confidence factor.