Executive summary
- P = Phenomenon, and Physical — the physics that produces the phenomenon.
- M = Mechanism, and the 4Ms of production: Man, Machine, Method, Material.
- It is for chronic losses that keep returning after repair, where Why-Why is too shallow.
- Eight steps, beginning with studying how the machine actually works — not with guessing causes.
- It complements the basic tools such as the 7 QC tools rather than replacing them.
Excellent plant performance is impossible while equipment reliability is short of complete — which is itself the consequence of incomplete maintainability. The usual cause is imperfect breakdown maintenance: a stop-gap fix with no search for the true root cause, either because nobody is looking or because nobody has the tool.
What P and M mean
| Symbol | Stands for | Meaning |
|---|---|---|
| P | Phenomenon | The event itself — the visible variation at the point where the loss occurs |
| P | Physical | Analysis to understand, in physical terms, how the machine works and why the phenomenon happens |
| M | Mechanism | The mechanism: what normal looks like, and what abnormal looks like |
| M | 4M | Man, Machine, Method, Material — the factors that create the conditions for the phenomenon |
The eight steps of P-M Analysis
- Study the machine’s basic principles How it was designed, its mechanisms and its structure. This is the foundation for every step that follows.
- Analyse the phenomenon physically Look at the event in terms of what is physically possible. This step decides whether the true cause will be found at all.
- List every possible condition Every condition that could produce the phenomenon, without pruning by intuition.
- Link conditions to the 4Ms Observe which of Man, Machine, Method and Material creates each of those conditions.
- Define the standard for each factor State what correct looks like — as a measurable number, not a description.
- Measure the actual condition against it Measure every factor on the machine and record the deviation. No memory, no opinion.
- Summarise the abnormalities found Collect every deviation and rank by impact.
- Correct every abnormal factor Fix all of them, verify with before-and-after data, then update the standard and the PM plan accordingly.
When to use Why-Why, when to use P-M
| Situation | The right tool |
|---|---|
| A one-off problem with a fairly direct cause | Why-Why analysis with the 7 QC tools |
| A repeating problem that is fixed successfully each time | Why-Why analysis pushed to system and standard level |
| Chronic loss that always returns after repair | P-M Analysis |
| Defects appearing intermittently with no clear pattern | P-M Analysis together with Quality Maintenance |
The most common mistake
Most teams start at step three by guessing conditions, skipping the study of how the machine works and the physical analysis. The result is a plausible-looking cause list that contains no actual root cause — and the problem returns two months later.