TY - JOUR
AU - Odougherty, Patrick
AB - Several changes in manufacturing environments over the last decade have created obstacles in correctly identifying the root cause of the ECU failure. Semiconductor manufacturers have improved their production processes and introduced new equipment to reduce particle counts which therefore decrease chances of electrical failures within their parts. Meanwhile, Tier I manufacturers have increased the ECU design complexity to provide more functionality in the vehicle and also increased the severity of their testing to weed out these failures that would reach the OEM customer or consumer market. The increase in ECU design complexity has resulted in more functional semiconductors such as microcontrollers, and large storage flash memory devices. All of these factors have led to struggles in failure analysis of the ECUs that result in non-supplier related defects. Using the historical production and failure data, various predictive models are developed and tested to predict whether a failure in ECU testing results in a supplier defect or a non-supplier defect. With an 81% of accuracy rate, results shift the focus away from the semiconductor supplier for non-related defects and toward solving the root cause which may be occurring within the testing or manufacturing process.
TI - A Predictive Analysis of Electronic Control Unit System Defects Within Automotive Manufacturing
JF - Journal of Failure Analysis and Prevention
DO - 10.1007/s11668-022-01401-0
DA - 2022-06-01
UR - https://www.deepdyve.com/lp/springer-journals/a-predictive-analysis-of-electronic-control-unit-system-defects-within-hFyams6fep
SP - 918
EP - 925
VL - 22
IS - 3
DP - DeepDyve
ER -