Failure Mode and Effects Table¶
Failure analysis table — instantiates Assumption Stress Testing
Adapts FMEA structure to assumption failure: one row per way a key premise could break, each rated for effect, severity, and detectability into a priority score, with a named mitigation.
A Failure Mode and Effects Table takes the premises a plan depends on and lays each one out as a row in a disciplined ledger: the assumption, the specific way it could break, the downstream effect if it does, a severity rating, a likelihood, a detectability rating (would anyone notice before the harm lands?), a resulting priority score, and the mitigation assigned to it. Borrowed from FMEA in reliability engineering, its one defining move is tabulation with a computed priority — it converts a scattered set of "what could go wrong" worries into a ranked, comparable, auditable list in which the failures that are severe and slow to detect rise to the top. It does not generate the assumptions in the first place, and it does not act them out in an unfolding future; it structures failure once the premises are already named, so effort flows to the modes that most deserve a safeguard.
Example¶
An automotive supplier is standing up a new production line for brake calipers, and the launch plan rests on a stack of quiet premises: that the incoming aluminum castings hold tolerance, that a single torque-station supplier ships on time, that operators reach rate within two weeks, that the leak-test rig catches every porous casting. The team builds a table. One row: "leak-test rig misses a slow porous casting." Effect — a caliper that weeps under thermal cycling, i.e., a field failure on a brake. Severity: near the top of the scale. Likelihood: low. Detectability: poor, because the current rig only pressure-holds for two seconds. Multiplied out, that row scores far higher than the mundane "operators ramp slowly" row despite being far less likely — precisely because a severe failure that hides is the dangerous kind. The mitigation column turns that number into an action: lengthen the pressure-hold and add a sampled teardown. The table's value is that it made an invisible, low-probability, high-consequence assumption failure out-rank the obvious ones, on paper, before the first shipment.
How it works¶
- One row per failure mode. Each key assumption is expanded into the concrete ways it could break — not "castings are fine" but "castings arrive out of tolerance" and, separately, "castings pass inspection but fail in service."
- Rate three axes. Severity of the effect, likelihood of the mode, and detectability — how likely you are to catch it before it does damage. Detectability is what separates this from a plain risk list: an undetected failure is worse than a visible one of equal severity.
- Compute a priority. Multiply the axes into a single Risk Priority Number so rows can be sorted. The score is a triage device, not a truth.
- Assign a mitigation and an owner per high-priority row, so the table exits as revised safeguards rather than as a document.
Tuning parameters¶
- Rating scale — a coarse 1–3 or a fine 1–10 per axis. Finer scales discriminate more but invite false precision and endless argument over a "6 versus 7."
- Action threshold — the priority score above which a mitigation is mandatory. Set it low and every row demands work; set it high and a genuinely dangerous mode slips under the bar.
- Detectability weighting — how heavily the hidden-failure axis counts. Weighting it up surfaces the silent modes this archetype most fears, at the cost of down-ranking loud, likely ones.
- Failure-mode granularity — one row per assumption or several. Splitting surfaces distinct effects but multiplies upkeep.
When it helps, and when it misleads¶
Its strength is that it forces every premise's failure into a comparable, documented, ranked form, and it uniquely rewards the failure you would not see coming — the row that is severe and undetectable. It leaves an audit trail a reviewer can challenge line by line.
Its failure mode is baked into the arithmetic: the priority score is built from ordinal ratings that are then multiplied and compared as if they were real quantities, so a tidy number can lend false authority to three subjective guesses.[n1] The classic misuse is scoring to the threshold — teams quietly rate a mode just under the action bar to avoid the mitigation work, turning the table into paperwork that certifies a plan rather than stresses it. The guarding discipline is to treat the priority number as a conversation-starter for triage, not a verdict: re-rate as real data arrives, and always eyeball the top severities directly regardless of where the multiplication lands them.
How it implements the components¶
failure_mode_map— the table is the map: each row traces one broken assumption through to its concrete downstream effect, degradation, or harm.criticality_filter— the priority score ranks the modes so scrutiny and mitigation are rationed to the severe-and-undetectable few rather than spread evenly.safeguard_revision— every high-priority row exits with an assigned mitigation, converting the analysis into a change to the plan's safeguards.
It does not surface or inventory the premises it tabulates (key_assumption, assumption_inventory) — that is Premortem — and it does not act the failures out in a constructed alternative future (stress_scenario, assumption_break_test) — that is Scenario Stress Test. This table structures failures that are already named.
Related¶
- Instantiates: Assumption Stress Testing — this table supplies the ranked failure-map-and-mitigation core of the archetype.
- Consumes: Premortem surfaces the candidate assumptions and failure stories this table then structures and rates.
- Sibling mechanisms: Premortem · Red-Team Future Challenge · Resilience Tabletop Exercise · Scenario Stress Test · Sensitivity Analysis Workshop · Stress-Test Scorecard · Trigger Dashboard
Editorial Notes¶
Form Classification¶
Form family: Analysis, Modeling & Optimization
Rationale: Failure Mode and Effects Table operates as a computation, comparison, model, or analytic representation used to infer, estimate, or choose because it adapts FMEA structure to assumption failure: one row per way a key premise could break, each rated for effect, severity, and detectability into a priority score, with a named mitigation.
Independent corroboration: The frozen evidence defines Failure Mode and Effects Table as 'Adapts FMEA structure to assumption failure: one row per way a key premise could break, each rated for effect, severity, and detectability into a priority score, with a named mitigation', so its operative form is Analysis, Modeling & Optimization.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Engineering & Design
Origin pattern: Single lineage
Present-day reach: Multi-domain
Rationale: The table directly adapts Failure Modes and Effects Analysis, a reliability-engineering method formalized in military systems work.
Related originating lineages:
- Military & Strategic Studies — Military equipment assurance materially shaped the early formal FMEA lineage.
- Organizational & Management Science — Risk registers and assumption governance materially shape ownership and mitigation tracking beyond physical engineering.
Review resolution: Both reviewers agree that engineering_design is primary. I retain military_strategic_studies, organizational_management only as formative origin lineages; single_lineage is appropriate because the alternate domains informed practice without constituting independent ownership. Reach is multi_domain because the structure transfers across several fields but is not a near-universal human pattern, an applicability judgment kept separate from provenance. Encyclopedia synthesis is true because the exact generalized artifact is an encyclopedia-authored combination or refinement. No unresolved historical ambiguity remains after reconciling the secondary fields.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] The Risk Priority Number (RPN = Severity × Occurrence × Detection) is the traditional FMEA priority score. Its well-known weakness is that the three inputs are ordinal ratings, so their product has no consistent meaning and equal RPNs can hide very different risk profiles — which is why current automotive FMEA guidance (AIAG-VDA) de-emphasizes the raw RPN in favor of an action-priority logic that looks at severity first. ↩