What it is
Redesign is the 5th and last Framework B outcome, alongside CBM, TBM, Proof Test and RTF: when NO proactive task is technically applicable and effective, and the failure consequence is not tolerable, RCM does not accept living with the risk — the way out is changing the design, not the maintenance routine. Redesign is not a synonym for "swap for a better piece of equipment": it means reassessing whether the CURRENT installation and required operating condition still match the original design assumptions — and, when they don't, redesigning.
When it is the right decision (Framework B)
Two simultaneous conditions (Moubray, ch. 9-10; SAE JA1011 §5.7):
- No proactive task passes the applicability/effectiveness test — CBM without a detectable P-F, TBM without a characteristic age (β ≈ 1), a proof test that would not reduce risk to the required level;
- The consequence is not tolerable — safety, environmental, or economic above the threshold RTF would accept.
At that intersection, RCM is categorical: there is no "accept the risk for now" — the design change is compulsory (it is the last line of Moubray's decision diagram).
There is no generic redesign — every installation is its own case
This is the point that usually gets lost: redesign is not applying a catalog recipe. It means reassessing the CURRENT condition of the installation and the required operation against the ORIGINAL design assumptions — does the design flow still match the actual operating flow? Has the fluid changed (viscosity, temperature, presence of solids)? Is the operating regime (starts/stops, turndown) still what the design assumed? Often the "chronic failure" that seems to call for redesign is actually a system operating outside the window it was designed for — and the correct redesign is specific to that installation, that fluid, that regime. There is no solution transferable from one plant to another without this reassessment.
Examples of what "redesigning" can mean
- Switching material/metallurgy to one more resistant to the identified failure mechanism;
- Adding redundancy (standby, N+1) when unavailability is the problem;
- Changing control/instrumentation logic to eliminate the operating mode that causes the failure;
- Resizing the equipment for the REAL operating condition — not the original design one, if it has changed.
On the platform
The strategy selector reaches redesign through RCM's two paths: protection that does not deliver the required availability (via Proof Test) and an evident failure with no applicable task and safety consequence (via RTF).