On Thu, Nov 8, 2018 Dennis Craggs presented “ASQ RRD Series: Big Data Analytics – Telematics Data Analysis”
Engineers conduct tests to verify that product meet engineering standards. These standards were developed with customer surveys, duplication past standards, or relying on expert opinions. Modern technology provides a new tool to validate standards by measuring product usage by the customer. An automotive example will be discussed, but the methods have much broader application.
Automobile companies installed telematics modules on fleet vehicles, with the consent of the owner, to collect and store usage and environmental data. The volume of data was enormous. Different analyic methods were required for different data types but there were only a few data types to consider. The raw data needed to be standardized for different vehicle miles or time. Then counting data, like trips/day, were analyzed with simple probability distributions. State data, like switch setting, need to be analyzed for transitions and time duration in states. Continuous measurements are more difficult. Engineers frequently use bar histograms, but bar histograms quickly become unreadable when two or more data sets were combined in the same graphic. Methods were developed that allowed many histograms to be analyzed fleet usage patterns and develop the 5th, 50th, and 95th customer percentiles. The standards can then be validated using the customer data.
Below a link with the slides of this webinar.