Reliability Programme

Problem

Instruments in the field needed an ongoing reliability programme to catch failures before they became a wider pattern, and the existing process of collecting and analysing that data was entirely manual, which meant it was slow and easy to fall behind on.

Process

I ran the reliability testing scheme day to day, creating test scripts and carrying out device maintenance. When something failed, I did root cause analysis using multimeters, oscilloscopes, manual measurement tools, and image dimensioning systems to work out what had actually gone wrong rather than just what had broken. I wrote Python scripts to automate the analysis of the resulting data sets, turning what had been a manual reporting job into something that ran on its own.

Outcome

The automation streamlined a testing process that had previously eaten a lot of manual hours, and the root cause analysis fed directly into a set of reliability improvements I suggested, including design changes and updated build instructions to stop assembly faults happening in the first place.