C3 IoT Proof of Technology: Vehicle Fleet Analytics
A global corporation in the transportation and logistics industry engaged in a 3 day Proof of Technology (PoT) around C3 IoT's Vehicle Fleet Analytics™ Product. The goal of the PoT was to demonstrate the ability to rapidly develop big data predictive analytic applications on the C3 IoT Platform, running on the AWS Cloud.
Organizations frequently implement a maintenance strategy for their fleets of vehicles using a combination of time and usage based maintenance schedules. While effective as a whole, time and usage based schedules do not take into account driving patterns, environmental factors, and sensors currently deployed within the vehicle measuring crank voltage, ignition voltage, and acceleration, all of which have a significant influence on the overall health of the vehicle.
In a typical fleet, a large percentage of road calls are related to electrical failure, with battery failure being a common cause. Battery failures result in unmet service agreement levels and costly re-adjustment of scheduled to provide replacement vehicles. To reduce the impact of unplanned maintenance, the transportation logistics company was interested in a trial of C3 Vehicle Fleet Analytics.
The PoT was built on roughly 1 TB of data from 10,000 vehicles, their maintenance records, and dozens of sensor measurements per vehicle. C3 IoT first defined the data model to store vehicle and vehicle sensor data. Using the metadata based C3 Type System, C3 IoT rapidly defined the data and canonical object models and the transformations required to convert source objects to the C3 data model. This then enabled the C3 IoT team to rapidly ingest the sensor data for all 10,000 vehicles in less than a day. With the data integrated, correlated, and normalized inside the C3 IoT Platform, the sensor data was then transformed into 26 distinct time series analytics which were fed into a machine learning classifier to predict battery failure.
With less than a week of work, the C3 IoT team:
- Developed a data and canonical objects model for vehicle operations
- Loaded a terabyte of vehicle and vehicle sensor data
- Developed 26 time series analytics
- Defined and executed a machine learning classifier across the entire data set to predict which vehicles would experience a battery failure
- Trained a machine learning classifier across the entire data set to predict battery failure