COMPRESSED-AIR FMEA, RISK & CRITICALITY ENGINEERING
Its SOURCED ON REQUEST capability state, SUPPLIER CONFIRMATION REQUIRED evidence state and Blower / Process-Air Engineer review boundary remain untouched.
The purpose of risk engineering is not to create a longer list of failures. it is to identify which failures matter, why they matter, and what must be controlled before they become incidents
It maps functions, failure modes, effects, causes, safeguards, detectability, recovery and criticality before the next outage occurs.
The page deliberately avoids pretending that one universal risk score can replace engineering judgment. Severity, occurrence, detectability and risk acceptance must follow the site's own governance, evidence and consequence framework.
1 — Define the System Function
Begin with what the compressed-air system must deliver: required flow, dynamic pressure, quality, availability and recovery at defined process boundaries.
2 — Define the Study Boundary
State whether the assessment covers source, treatment, storage, distribution, controls, utilities, end use and emergency supply.
3 — Define the Operating States
Normal, low load, peak production, maintenance, startup, shutdown, degraded operation and emergency supply can have different failure exposure.
4 — Define Critical Loads
Risk analysis must use that service definition.
5 — FMEA Is a Structured Question Set
For each function ask: how can it fail, what happens, why can it happen, how is it detected, what controls exist and what action reduces risk.
6 — Function Before Component
A dryer is not merely an asset; its function is to deliver the required air quality at the required flow and pressure.
7 — Failure Mode
Describe how the function is lost or degraded: fails to start, low capacity, high pressure drop, wet air, communication loss or similar.
8 — Failure Effect
Describe the consequence at the next system/process boundary rather than only the component symptom.
9 — Failure Cause
Identify supported physical, control, maintenance, utility, environmental or human-system causes.
10 — Existing Controls
Record prevention, detection, redundancy, alarms, inspection, storage, standby and recovery controls already present.
11 — Severity
Evaluate consequence using site-defined safety, production, quality, environmental and reliability criteria.
12 — Occurrence
Use actual history, condition evidence or defensible engineering judgment; do not invent failure probabilities.
13 — Detectability
Consider whether degradation is visible before consequence occurs and whether alarms/sensors are trustworthy.
14 — Risk Priority Numbers Have Limits
Multiplying ordinal scores can help prioritisation but can hide high-severity items. Never allow one arithmetic score to override consequence judgment.
15 — Criticality Can Be Qualitative
High/medium/low or another controlled framework can be valid when quantitative failure data are weak.
16 — Risk Acceptance Is Organisational
Apply the site's approved governance.
17 — Source Compressor Failure Mode
Loss of capacity can produce pressure decay, additional starts, production interruption or loss of redundancy.
18 — Compressor Degraded Capacity
A machine can remain running while delivering less corrected airflow; this is a hidden failure mode.
19 — Compressor High-Temperature Trip
Potential causes span cooling, ventilation, ambient, fouling, water supply and package condition.
20 — Compressor Control Failure
A healthy mechanical package can become unavailable through sensor, controller, drive or communication faults.
21 — Standby Failure Mode
Standby can fail because it is faulted, isolated, unmaintained, lacks utilities or cannot start within the required transition.
22 — Treatment Failure Mode
Dryer failure can preserve pressure while silently violating air-quality requirements.
23 — Filter Failure Mode
Excessive differential can starve users; damaged/bypassed filtration can compromise quality.
24 — Drain Failure Mode
Failed-open drains waste capacity; failed-closed drains increase condensate risk.
25 — Receiver Failure / Unavailability
Isolation, drainage or condition issues can remove transient storage from the active system.
26 — Distribution Failure Mode
Rupture, corrosion, valve closure, restriction or poor topology can interrupt one or multiple zones.
27 — Local Connection Failure
Hoses, couplings, regulators and valves can create machine-level starvation while the main header appears healthy.
28 — Master Controller Failure
Central optimisation can fail while local fallback preserves service—or fail dangerously if fallback was never engineered.
29 — Pressure Sensor Failure
Bias, drift or loss can cause incorrect fleet response and source pressure.
30 — Flow Meter Failure
Bad flow data can corrupt capacity, energy and leak decisions without immediately stopping production.
31 — Dew-Point Sensor Failure
False dry/wet indication can either waste dryer energy or conceal quality failure.
32 — Electrical Supply Failure
One feeder can remove several nominally redundant compressors and treatment assets.
33 — Ventilation Failure
One fan/louver path can become a common-mode thermal failure.
34 — Cooling-Water Failure
Shared pumps, towers or water supply can remove multiple water-cooled machines.
35 — Communication Network Failure
A common switch/network can remove coordinated control or monitoring across the fleet.
36 — Human / Configuration Failure
Wrong valve position, manual override or undocumented setpoint can defeat otherwise sound architecture.
37 — Maintenance-Induced Failure
Incorrect assembly, contamination, wrong parts or settings can create new failure after service.
38 — Expansion-Induced Failure
New production can consume capacity, storage, treatment or pressure margin without any component failing.
39 — Environmental Failure Driver
Dust, heat, corrosive vapour, flooding or water quality can alter equipment reliability.
40 — Supplier / Obsolescence Risk
Long lead times or unsupported controls can extend recovery even when the physical failure is straightforward.
41 — Single-Point-of-Failure Analysis
Identify one component or dependency whose loss removes the defined critical service.
42 — Common-Mode Analysis
Identify one cause that defeats multiple redundant components simultaneously.
43 — Dependency Analysis
Map each critical load to source, treatment, storage, controls, electrical/cooling utilities and distribution.
44 — Barrier Analysis
For each important failure, identify preventive, detective, mitigating and recovery barriers.
45 — Barrier Independence
Two safeguards relying on the same sensor, power source or controller may not be truly independent.
46 — Barrier Health
A safeguard that exists on a drawing but is bypassed, faulted or untested is not a reliable control.
47 — Proof Testing
Standby, alarms, fallback control and emergency connections may require periodic proof appropriate to consequence.
48 — Detection Latency
A slowly detected leak or dryer failure can create larger consequence than one detected immediately.
49 — Recovery Latency
Spares, access, competence, rental lead time and restart sequence influence how long service remains degraded.
51 — Emergency Supply as a Barrier
A rental tie-in is useful only when capacity, treatment, connection and mobilisation time are compatible with the tolerated outage.
52 — Load Shedding as a Barrier
Approved non-critical load shedding can preserve critical pressure during degraded operation.
54 — Excess Capacity Is Not Automatically Low Risk
Oversized machines can cycle inefficiently and still share common utilities.
55 — Redundancy Without Independence
Two compressors on one feeder or one dryer remain vulnerable to common failure.
56 — Worked FMEA: Compressor Trip
Illustrative only: duty compressor trips at peak load. Effect depends on standby readiness, storage bridge, staging delay and remaining corrected capacity.
57 — Worked FMEA: Dryer Failure
Illustrative only: pressure remains normal while dew point deteriorates. Quality-sensitive production can fail even though compressor capacity is adequate.
58 — Worked FMEA: Ventilation Failure
Illustrative only: one exhaust fan failure raises room temperature and several compressors trip. The fan is a common-mode critical asset.
59 — Worked FMEA: Header Isolation
Illustrative only: a normally open valve is closed after maintenance and one zone starves. Labeling, valve-state verification and commissioning become controls.
60 — Worked FMEA: Sensor Drift
Illustrative only: master pressure sensor reads low, raising plant pressure and energy while no alarm occurs. Sensor validation becomes a detection barrier.
61 — Worked FMEA: Drain Failure
Illustrative only: a failed-open drain increases background demand and brings another compressor online. The energy symptom originates in maintenance.
62 — Worked FMEA: New Production
Illustrative only: added pneumatic equipment consumes previous standby margin. Normal production works but N+1 no longer passes.
63 — Risk Reduction by Elimination
Where practical, remove unnecessary demand, obsolete branches or avoidable failure exposure.
64 — Risk Reduction by Design
Add segmentation, independent utilities, storage, redundancy or better access where consequence justifies it.
65 — Risk Reduction by Controls
Use alarms, fallback logic, interlocks and monitoring where appropriate, while preserving safe failure modes.
66 — Risk Reduction by Maintenance
Condition monitoring, preventive tasks and critical spares can reduce occurrence or recovery time.
67 — Risk Reduction by Procedure
Procedures can reduce configuration and recovery errors but should not compensate indefinitely for poor engineering.
68 — Risk Reduction by Training
Training supports correct response when a verified competence gap exists.
69 — Risk Reduction by Instrumentation
Measurement can improve detectability, but only if sensors are maintained and alarms are acted upon.
70 — Risk Reduction by Proof Testing
Testing validates that standby, alarms and recovery paths actually work.
77 — Risk Register
Maintain hazard/failure, effect, cause, existing controls, rating, action, owner and residual risk.
78 — FMEA Register
Maintain function, failure mode, local/system effect, cause, detection, controls and recommended action.
79 — Criticality Register
Identify assets whose failure consequence or recovery burden justifies stronger maintenance/spares.
80 — Single-Point Register
Track unresolved components whose loss defeats critical service.
81 — Common-Mode Register
Track shared utilities, controls and environmental dependencies.
82 — Barrier Register
Track each important safeguard, owner, proof-test method and current status.
83 — Residual-Risk Register
After actions, record what risk remains and why it is accepted or still open.
84 — Action Register
Record action, owner, due date, status, verification and effectiveness.
85 — Assumption Register
Record uncertain demand, failure, recovery or supplier assumptions that materially affect conclusions.
86 — Evidence Register
Link drawings, logs, proof tests, incident history, supplier data and calculations.
87 — Risk Review After Expansion
New demand can change severity, occurrence and recovery capability.
88 — Risk Review After Compressor Replacement
Capacity, controls, heat rejection and spares change.
89 — Risk Review After Dryer Replacement
Quality, pressure drop, utilities and common-mode treatment exposure change.
90 — Risk Review After Control Upgrade
Centralisation can improve efficiency while introducing new common dependencies.
91 — Risk Review After Distribution Change
New valves, ring closures or branches can change isolation and alternate-flow paths.
92 — Risk Review After Pressure Change
Lower pressure can reduce demand but change storage and regulator margin.
93 — Risk Review After Leak Campaign
Released capacity can improve resilience until reconsumed by production.
95 — Risk Review After Major Maintenance
Valve states, standby availability and configurations should be reconfirmed.
96 — Risk Review After Long Shutdown
Restart, drains, controls, treatment and utilities can have different failure exposure.
97 — Risk Review Frequency
Use change, incident and criticality triggers rather than one universal calendar interval.
98 — Dynamic Risk
A system can be low risk in normal operation and high risk during maintenance when standby/treatment is unavailable.
99 — Maintenance Risk Window
Before planned work, recalculate remaining capacity, treatment and common dependencies.
100 — Temporary Configuration Risk
Rental hoses, bypasses and manual modes require explicit temporary risk review.
101 — Startup Risk
Simultaneous compressor/process restart can create electrical and pneumatic transients.
102 — Shutdown Risk
Poor shutdown sequence can leave trapped pressure, wet piping or unavailable treatment on restart.
103 — Black-Start Risk
Loss of power can remove ventilation, dryers, controls and compressors together; recovery order matters.
104 — Cyber / Control Risk
Connected control failure, access or configuration loss can affect availability; follow site IT/OT governance.
105 — Obsolescence Risk
Unsupported controllers or drives can create long recovery even with healthy mechanical assets.
106 — Spare-Part Risk
A critical spare with long lead time can dominate MTTR and operational availability.
107 — Competence Risk
A technically recoverable failure can become prolonged if required specialist skills are unavailable.
108 — Documentation Risk
Wrong or missing drawings/settings can increase diagnostic and restoration error.
109 — Contractor Risk
Temporary work can create undocumented valve/control states if handover is weak.
110 — Procurement Risk
Buying equipment without verified pressure, flow, quality, controls and service evidence can embed future failure modes.
111 — Risk and Energy
Efficiency actions must retain pressure, quality, standby and recovery guardrails.
125 — Risk and Emergency Air
Emergency supply must be treated as an engineered barrier with capacity, quality and mobilisation evidence.
126 — Quantitative Analysis Has a Place
Where credible failure-rate and consequence data exist, quantitative reliability/risk methods can support decisions.
127 — Quantitative Analysis Has Limits
Sparse or non-comparable data can create false precision. State confidence and assumptions.
128 — Sensitivity Analysis
Test whether the priority changes when uncertain demand, repair time or occurrence assumptions vary.
129 — Scenario Analysis
Use credible normal, peak, maintenance and outage scenarios rather than one operating point.
130 — Bow-Tie Thinking
For high-consequence top events, map threats and preventive barriers on one side and mitigating/recovery barriers on the other.
131 — Barrier Ownership
Every critical barrier needs a responsible owner and health status.
132 — Barrier Degradation
A bypassed alarm, overdue proof test or unavailable spare can increase risk without any new equipment failure.
133 — Barrier Independence Review
Confirm redundant barriers do not secretly share power, sensor, controller or human action.
134 — Risk-Based Maintenance
Use criticality and failure modes to focus maintenance effort where it changes consequence or occurrence.
135 — Risk-Based Spares
Stock decisions should reflect failure consequence, lead time, interchangeability and recovery strategy.
136 — Risk-Based Instrumentation
Add sensors where improved detectability materially reduces consequence or diagnostic delay.
137 — Risk-Based Redundancy
Add redundant source, treatment, controls or utilities only when the service consequence justifies lifecycle cost/complexity.
138 — Risk-Based Testing
Proof-test frequency and depth should follow consequence, failure mechanism and site governance.
139 — Risk-Based Training
Focus competence on recovery tasks and failure modes with meaningful consequence.
140 — Risk-Based Documentation
Critical recovery drawings, valve maps and settings deserve stronger configuration control.
141 — Risk Communication
Present the failure mode, consequence, evidence, current controls, residual risk and action—not only a colour or score.
142 — Executive Risk View
Management needs unresolved high-consequence single points, overdue actions, standby readiness and investment decisions.
143 — Engineering Risk View
Engineers need detailed failure modes, dependencies, barrier health, assumptions and verification.
144 — Operations Risk View
Operators need approved normal/degraded modes, alarms, isolation boundaries and escalation.
145 — Maintenance Risk View
Maintenance needs critical tasks, spares, proof tests and defect priorities.
146 — Change-Control Link
No material system modification should close until relevant FMEA, criticality and barrier records are updated.
147 — Auditability
A competent reviewer should be able to trace why an item was considered critical and what evidence supports the action.