Mastering Preventive & Condition‑Based Maintenance: A Practical PM Development Guide

Many organizations chase the latest buzzwords without laying the foundation to support them. Preventive maintenance (PM) and condition‑based maintenance (CBM) are classic examples. This guide is not exhaustive but serves as a starting point for conversations about building a robust PM program.

1. Eliminate non‑value‑added tasks and steps from your PMs. Time is rarely available for task updates, yet we continue performing the wrong work.
2. Focus first on equipment that is costing you most—often the non‑critical assets that bleed resources.
3. Identify equipment failure modes. For example, the Gulf oil spill blow‑out preventer had over 260 failure modes; SKF cites more than 50 per bearing.
4. Apply the Reliability‑Centered Maintenance (RCM) logic tree to select tasks. Prioritize condition‑based monitoring (CBM) whenever possible, as it is typically less costly than full overhauls. Some teams still perform overhauls simply because it's the historical norm. Your equipment dictates the appropriate PM type; run‑to‑failure can be acceptable if collateral damage is controlled. CBM extends the failure detection window compared to time‑based schedules.
5. Define tasks that detect each failure mode early. Failure‑Mode and Effects Analysis (FMEA) is invaluable, yet many groups cannot allocate time to develop procedures.
6. Document procedures using standardized, clear language.
7. Emphasize precision: adhere to specifications such as torque values and belt tension.
8. Seek hidden failures that may not surface during routine checks.
9. Set task frequency based on data—ideally half the width of the failure detection window—while ensuring you can plan and schedule maintenance proactively.
10. Inspect and audit the work processes regularly.
11. Measure outcomes and benefits quantitatively.
12. Gather feedback from technicians and incorporate a continuous‑improvement loop.
13. Good procedures alone are insufficient; change worker behavior, which can account for up to 80% of maintenance effectiveness.
14. Build partnerships with production, engineering, and other stakeholders. PM is ineffective if production denies access to equipment. Even well‑maintained gear can fail if operated incorrectly or installed poorly. Remember that total life‑cycle cost can exceed 20× the initial purchase price.
15. After a failure, conduct a root‑cause (autopsy) analysis to determine why the failure went undetected. Review task procedures and frequencies, then update the PM accordingly.