Systems and Methods of Determining Effectiveness of Vehicle Safety Features
Systems and methods for determining the effectiveness of vehicle safety features are provided. Vehicle data is obtained for vehicles having various smart safety features, and a list of translated vehicle build records is generated from the obtained data applying OEM-agnostic terminology for the smart safety features. A machine learning algorithm may be trained to generate an effectiveness score associated with one or more smart safety features, at least by analyzing a plurality of translated vehicle build records, vehicle telematics data, and vehicle accident records associated with each of the plurality of vehicles.
1 . A computer-implemented method for determining effectiveness of vehicle safety features in a manner agnostic to terminology used by original equipment manufacturers (OEMs), the computer-implemented method comprising:
generating, by one or more processors, a plurality of translated vehicle build records for a plurality of vehicles each having one or more smart safety features by applying an ontology model to a corresponding plurality of vehicle-specific build records of the plurality of vehicles, wherein the ontology model maps OEM-specific terminology for each of a plurality of smart safety features to OEM-agnostic terminology for the respective smart safety feature;
obtaining, by the one or more processors from a plurality of on-board computing devices associated with the plurality of vehicles, usage data indicating an extent of usage of the smart safety features of each respective vehicle of the plurality of vehicles at a plurality of times during operation of the respective vehicle;
obtaining, by the one or more processors, vehicle accident record information for a plurality of vehicle accidents involving at least some of the plurality of vehicles, wherein the vehicle accident record information includes indications of times of the plurality of vehicle accidents;
determining, by the one or more processors, smart safety feature usage data associated with the plurality of vehicle accidents based upon the plurality of translated vehicle build records, the usage data, and the vehicle accident record information; and
generating, by the one or more processors, an effectiveness score associated with each smart safety feature in each of a plurality of environments by analyzing the smart safety feature usage data and the vehicle accident record information.
2 . The computer-implemented method of claim 1 , wherein the computer-implemented method further comprises:
obtaining, by the one or more processors, the vehicle-specific build records for the plurality of vehicles from a vehicle feature database associating each Vehicle Identification Number (VIN) with installed smart safety features.
3 . The computer-implemented method of claim 1 , wherein the computer-implemented method further comprises:
generating, by the one or more processors, the ontology model by training a machine learning model using OEM-specific terminology associated with each of the plurality smart safety features for a plurality of OEMs.
4 . The computer-implemented method of claim 1 , wherein the computer-implemented method further comprises:
obtaining, by the one or more processors from the plurality of on-board computing devices associated with the plurality of vehicles, vehicle telematics data for each of the plurality of vehicles, wherein the vehicle telematics data includes sensor data from one or more sensors disposed at each respective vehicle; and
determining, by the one or more processors, an operating environment in which each respective vehicle of the plurality of vehicles was operating at a plurality of times during operation of the respective vehicle,
wherein generating the effectiveness score associated with each smart safety feature in each of the plurality of environments is further based upon the determined operating environments.
5 . The computer-implemented method of claim 1 , wherein the vehicle accident record information includes information regarding a severity of each of the plurality of vehicle accidents.
6 . The computer-implemented method of claim 1 , wherein the vehicle accident record information includes information regarding a frequency of accidents for each of one or more combinations of the smart safety features.
7 . The computer-implemented method of claim 1 , wherein generating the effectiveness score associated with each smart safety feature comprises predicting the effectiveness score of the respective smart safety feature in combination with at least one other smart safety feature.
8 . The computer-implemented method of claim 1 , wherein the computer-implemented method further comprises:
obtaining, by the one or more processors, vehicle information regarding a subset of the plurality of smart safety features installed in a particular vehicle of the plurality of vehicles; and
determining, by the one or more processors, an insurance rating for the particular vehicle based upon the effectiveness scores associated with each smart safety feature of the subset of smart safety features.
9 . A computer system for determining effectiveness of vehicle safety features in a manner agnostic to terminology used by original equipment manufacturers (OEMs), the computer system comprising:
one or more processors; and
one or more non-transitory program memories communicatively coupled to the one or more processors and storing executable instructions that, when executed by the one or more processors, cause the computer system to:
generate a plurality of translated vehicle build records for a plurality of vehicles each having one or more smart safety features by applying an ontology model to a corresponding plurality of vehicle-specific build records of the plurality of vehicles, wherein the ontology model maps OEM-specific terminology for each of a plurality of smart safety features to OEM-agnostic terminology for the respective smart safety feature;
obtain, from a plurality of on-board computing devices associated with the plurality of vehicles, usage data indicating an extent of usage of the smart safety features of each respective vehicle of the plurality of vehicles at a plurality of times during operation of the respective vehicle;
obtain vehicle accident record information for a plurality of vehicle accidents involving at least some of the plurality of vehicles, wherein the vehicle accident record information includes indications of times of the plurality of vehicle accidents;
determine smart safety feature usage data associated with the plurality of vehicle accidents based upon the plurality of translated vehicle build records, the usage data, and the vehicle accident record information; and
generate an effectiveness score associated with each smart safety feature in each of a plurality of environments by analyzing the smart safety feature usage data and the vehicle accident record information.
10 . The computer system of claim 9 , wherein the executable instructions further cause the computer system to:
obtain the vehicle-specific build records for the plurality of vehicles from a vehicle feature database associating each Vehicle Identification Number (VIN) with installed smart safety features.
11 . The computer system of claim 9 , wherein the executable instructions further cause the computer system to:
generate the ontology model by training a machine learning model using OEM-specific terminology associated with each of the plurality smart safety features for a plurality of OEMs.
12 . The computer system of claim 9 , wherein:
the executable instructions further cause the computer system to:
obtain, from the plurality of on-board computing devices associated with the plurality of vehicles, vehicle telematics data for each of the plurality of vehicles, wherein the vehicle telematics data includes sensor data from one or more sensors disposed at each respective vehicle; and
determine an operating environment in which each respective vehicle of the plurality of vehicles was operating at a plurality of times during operation of the respective vehicle; and
the executable instructions that cause the computer system to generate the effectiveness score associated with each smart safety feature in each of the plurality of environments cause the computer system to generate the effectiveness score based upon the determined operating environments.
13 . The computer system of claim 9 , wherein the vehicle accident record information includes information regarding a severity of each of the plurality of vehicle accidents.
14 . The computer system of claim 9 , the executable instructions that cause the computer system to generate the effectiveness score associated with each smart safety feature in each of the plurality of environments cause the computer system to predict the effectiveness score of the respective smart safety feature in combination with at least one other smart safety feature.
15 . A tangible, non-transitory computer-readable storage medium storing computer-executable instructions for determining effectiveness of vehicle safety features in a manner agnostic to terminology used by original equipment manufacturers (OEMs) that, when executed by at least one processor of a computer system, cause the computer system to:
generate a plurality of translated vehicle build records for a plurality of vehicles each having one or more smart safety features by applying an ontology model to a corresponding plurality of vehicle-specific build records of the plurality of vehicles, wherein the ontology model maps OEM-specific terminology for each of a plurality of smart safety features to OEM-agnostic terminology for the respective smart safety feature;
obtain, from a plurality of on-board computing devices associated with the plurality of vehicles, usage data indicating an extent of usage of the smart safety features of each respective vehicle of the plurality of vehicles at a plurality of times during operation of the respective vehicle;
obtain vehicle accident record information for a plurality of vehicle accidents involving at least some of the plurality of vehicles, wherein the vehicle accident record information includes indications of times of the plurality of vehicle accidents;
determine smart safety feature usage data associated with the plurality of vehicle accidents based upon the plurality of translated vehicle build records, the usage data, and the vehicle accident record information; and
generate an effectiveness score associated with each smart safety feature in each of a plurality of environments by analyzing the smart safety feature usage data and the vehicle accident record information.
16 . The tangible, non-transitory computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
obtain the vehicle-specific build records for the plurality of vehicles from a vehicle feature database associating each Vehicle Identification Number (VIN) with installed smart safety features.
17 . The tangible, non-transitory computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
generate the ontology model by training a machine learning model using OEM-specific terminology associated with each of the plurality smart safety features for a plurality of OEMs.
18 . The tangible, non-transitory computer-readable storage medium of claim 15 , wherein:
the executable instructions further cause the computer system to:
obtain, from the plurality of on-board computing devices associated with the plurality of vehicles, vehicle telematics data for each of the plurality of vehicles, wherein the vehicle telematics data includes sensor data from one or more sensors disposed at each respective vehicle; and
determine an operating environment in which each respective vehicle of the plurality of vehicles was operating at a plurality of times during operation of the respective vehicle; and
the executable instructions that cause the computer system to generate the effectiveness score associated with each smart safety feature in each of the plurality of environments cause the computer system to generate the effectiveness score based upon the determined operating environments.
19 . The tangible, non-transitory computer-readable storage medium of claim 15 , wherein the vehicle accident record information includes information regarding a severity of each of the plurality of vehicle accidents.
20 . The tangible, non-transitory computer-readable storage medium of claim 15 , the executable instructions that cause the computer system to generate the effectiveness score associated with each smart safety feature in each of the plurality of environments cause the computer system to predict the effectiveness score of the respective smart safety feature in combination with at least one other smart safety feature.