Failure Time Analysis Applied to Boeing 737 MAX

Jan B. Smith

DOI: https://doi.org/10.1109/rams48030.2020.9153706

2020-01-01

Abstract:Data for the analysis comes from Boeing publications and public announcements. The analysis input data are believed to be conservative, i.e., the data produces results that tend to understate the significance of the analysis findings and conclusions. Using data existing at the time, the analysis shows the 737 MAX should have been grounded sooner. Perhaps after the first event (crash) but certainly after the second. The decision not to ground after the second event led to high risk of a third event. If all planes remained in service, the risk of a third event in three days is about 1 in 25. This high risk indicates that Boeing and the FAA did not do a failure time analysis. It is supposed those that chose to voluntary ground their aircraft made the decision without data analysis as well. At the time of the events, cause and effect data were sparse, but the event time data were readily and immediately available. This unused data were of tremendous value in quantifying risk for decision making. An objective of this paper is to assure failure time data are analyzed in the future, not only in engineered systems, but in business, medical, and any other data system where special cause events are of interest and a homogeneous Poisson process (HPP) is a reasonable null hypothesis.