Tidy Data for Effective Data Science – with Application to Flying Lead Reliability

Project Description

Data science is being increasingly acknowledged for its potential to aid decision making in engineering analysis. Particularly in the resources industries, there is potentially much to gain by increasing collaboration between data owners, who have significant application experience, and data analysts, who are able to implement modern data science techniques

A Bayesian statistical model was developed to better predict failure time and location of flying leads, enabling streamlined rectification works. General recommendations were also made on statistical modelling and data acquisition.

Fig 1:  Effect of Flying Lead Length

Note:  Note that the blue points represent that knowledge that the asset has lasted for at least that long, only the red data are true failure times.  There does not appear to be a correlation between length and failure times.