Using constantly updated supplier data and historical data from their own repairs, we used a predictive data science model to help this manufacturer identify circuit board components that are most likely to have been incorrectly installed based on the fault presented.
