Apparatus and method for processing data to improve operation of motor vehicles

1 . A method comprising:

receiving a set of current data values for a group of vehicles over a network, the group of vehicles operating in unison as a vehicle fleet, each of the current data values being generated by at least one of the vehicles in the group of vehicles;

retrieving data external to the group of vehicles from at least one database;

processing the set of current data values and the received external data for the group of vehicles using a selected machine learning algorithm from a set of at least two machine learning algorithms to identify a subset of selected processed data values, the processing further including:

evaluating the processing of the set of current data values by each machine learning algorithm from the set of at least two machine learning algorithms for accuracy against a target vehicle operational outcome based on a threshold confidence interval;

determining if at least one of machine learning algorithms from the set of at least two machine learning algorithms is stable and unbiased if the accuracy of the at least one of machine learning algorithms against the target vehicle operational outcome exceeds the threshold confidence interval;

selecting one of the at least one machine learning algorithms as the selected machine learning algorithm to generate a score value for each of the processed set of current data values and received external data using a real time adjustability factor if it is determined that at least one of the machine learning algorithms is stable and unbiased, the real time adjustability factor identifying how each of the processed set of current data values and received external data contribute to the target vehicle operational outcome; and

selecting a subset of the processed set of current data values and received external data having the highest score value;

processing the set of selected data values, using the selected machine learning algorithm, to generate a current expected efficiency value for each vehicle in the group of vehicles;

processing the current expected efficiency value for the group of vehicles, using a second machine learning algorithm, to generate at least one predicted efficiency value at a time in the future for each vehicle in the group of vehicles at a future point in time;

determining if one of the current expected efficiency value and the at least one predicted efficiency value at a time in the future for a first vehicle in the group of vehicles is less than one of the current expected efficiency value and the corresponding at least one predicted efficiency value at a time in the future for a second vehicle in the group of vehicles;

comparing at least one operational characteristic of the first vehicle and the second vehicle if it is determined that one of the current expected efficiency value and the at least one predicted efficiency value at a time in the future for the first vehicle is less than one of the current expected efficiency value and the corresponding at least one predicted efficiency value at a time in the future for the second vehicle; and

providing a message in order to improve efficiency for the first vehicle in the group of vehicles based on the comparison.

2 . The method of claim 1 , wherein the current expected efficiency is at least one of current expected fuel efficiency and current expected battery efficiency and the at least one future predicted efficiency value at a time in the future is at least one of predicted fuel efficiency at a time in the future and predicted battery efficiency at a time in the future.

3 . The method of claim 1 , wherein the efficiency improvement includes at least one of modifying a driving characteristic of the first vehicle, adjusting a travel route of the first vehicle, and identifying maintenance for the first vehicle.

4 . The method of claim 3 , wherein the at least one operational characteristic of the first vehicle is at least one of vehicle acceleration, vehicle operation idle time, and sudden vehicle braking.

5 . The method of claim 3 , wherein the modifying of the driving of the first vehicle includes adjusting a speed of the first vehicle automatically by sending control instructions to the first vehicle over a network.

6 . The method of claim 1 , wherein providing the message includes providing the message to an operator of the at least one vehicle while driving the first vehicle.

7 . The method of claim 1 , wherein the selected machine learning algorithm is not the same as the second machine learning algorithm.

8 . The method of claim 1 , wherein the data external to the group of vehicles includes at least one of geographic location for a vehicle in the group of vehicles, vehicle type identification for a vehicle in the group of vehicles, age of a vehicle, and environmental conditions around a vehicle in the group of vehicles.

9 . The method of claim 1 , wherein the current set of data values are provided by a set of sensors included in each one of the vehicles in the group of vehicles.

10 . A system comprising:

a data input element, the data input element receiving a set of current data values for a group of vehicles over a network, the group of vehicles operating in unison as a vehicle fleet, each of the current data values being generated by at least one of the vehicles in the group of vehicles, the data input element further retrieving data external to the group of vehicles from at least one database;

an artificial neural network engine coupled to the data input element, the artificial neural network engine processing the set of current data values and the received external data for the group of vehicles using a selected machine learning algorithm from a set of at least two machine learning algorithms to identify a subset of selected processed data values, the processing further including evaluating the processing of the set of current data values by each machine learning algorithm from the set of at least two machine learning algorithms for accuracy against a target vehicle operational outcome based on a threshold confidence interval, determining if at least one of machine learning algorithms from the set of at least two machine learning algorithms is stable and unbiased if the accuracy of the at least one of machine learning algorithms against the target vehicle operational outcome exceeds the threshold confidence interval, selecting one of the at least one machine learning algorithms as the selected machine learning algorithm to generate a score value for each of the processed set of current data values and received external data using a real time adjustability factor if it is determined that at least one of the machine learning algorithms is stable and unbiased, the real time adjustability factor identifying bow each of the processed set of current data values and received external data contribute to the target vehicle operational outcome, and selecting a subset of the processed set of current data values and received external data having the highest score value, the artificial neural network engine further processing the set of selected data values, using the selected machine learning algorithm, to generate a current expected efficiency value for each vehicle in the group of vehicles and processing the current predicted efficiency value for the group of vehicles, using a second machine learning algorithm, to generate at least one predicted efficiency value at a time in the future for each vehicle in the group of vehicles;

a data processing device coupled to the artificial neural network engine, the data processing device determining if one of the current expected efficiency value and the at least one predicted efficiency value at a time in the future for a first vehicle in the group of vehicles is less than one of the current expected efficiency value and the corresponding at least one future predicted efficiency value at a time in the future for a second vehicle in the group of vehicles and comparing at least one operational characteristic of the first vehicle and the second vehicle if it is determined that one of the current expected efficiency value and the at least one predicted efficiency value at a time in the future for the first vehicle is less than one of the current expected efficiency value and the corresponding at least one predicted efficiency value at a time in the future for the second vehicle; and

a data output element coupled to the artificial neural network engine and the data processing device, the data output element providing a message in order to improve efficiency for the first vehicle based on the comparison.

11 . The system of claim 10 , wherein the current pe efficiency is at least one of current expected fuel efficiency and current expected battery efficiency and the at least one predicted efficiency value at a time in the future is at least one of predicted fuel efficiency at a time in the future and predicted battery efficiency at a time in the future.

12 . The system of claim 10 , wherein the efficiency improvement includes at least one of modifying a driving characteristic of the first vehicle, adjusting a travel route of the first vehicle, and identifying maintenance for the first vehicle.

13 . The system of claim 12 , wherein the at least one driving characteristic is at least one of vehicle acceleration, vehicle operation idle time, and vehicle braking.

14 . The system of claim 12 , wherein the data output element provides the message to an operator of the at least one vehicle while driving the first vehicle.

15 . The system of claim 11 , wherein the modifying of the driving of the first vehicle includes adjusting a speed of the first vehicle automatically by sending control instructions to the first vehicle over a network.

16 . The system of claim 10 , wherein the selected machine learning algorithm is not the same as the second machine learning algorithm.