Problem
From 2021 to 2023, as a reliability data analyst at Pinnacle, Inc., I worked on the reliability program for one of the world's largest open-pit copper mines, covering its mining, concentrating, smelting and refining plants. Equipment failure there means lost production, and for piping and vessels it can also mean loss of containment. Inspection effort has to go where the risk is.
Constraints
- Four facilities, each with its own plant leadership and its own equipment.
- The programs had to run inside the asset-management systems the site already used: SAP and GE APM.
- Each program had to be put in place together with that plant's leadership.
What I built
I implemented risk-based inspection (RBI), failure modes and effects analysis (FMEA) and reliability-centred maintenance (RCM) programs.
I analysed equipment corrosion rates, failure modes, consequences and likelihood of failure, and predicted equipment lifespans. That risk ranking set the routine inspection tactics, which I created, tracked and scheduled in SAP and GE APM. I used Python to pull usable findings out of mechanical and process data, and built machine-learning workflows that sped up document processing and data transformation. I also qualified as a Level II NDT technician during the work.
How it was verified
I coordinated with plant leadership at each of the four facilities to put validated routine inspection programs in place. The downtime figure below is an estimate of potential downtime avoided, not a measured saving.
Outcome
Est. $30M+ in potential downtime avoided through the redesigned maintenance program.
Stack
- RBI
- FMEA
- RCM
- SAP
- GE APM
- Python