10.2 Post-Work Documentation & Failure History Capture

Post-Work Documentation & Failure History Capture

Quick Answer: Work closeout should preserve what asset and location were involved, what was found and done, labor and materials, as-found/as-left measurements, tests, restored condition, relevant failure information, and needed follow-up. The record supports operations and engineering and may also satisfy specific regulatory, contractual, warranty, or legal duties.

---

The Strategic Importance of Work Order Closeout (SMRP BoK 5.6)

In immature maintenance cultures, technicians and supervisors view the work order as a bureaucratic hurdle. Once the physical machine is running, work orders are often hastily marked "done" or left open for weeks. This administrative neglect creates "dark data"—leaving the plant completely blind to the actual performance, failure modes, and true life-cycle costs of its capital assets.

Work order closeout is an essential phase of the asset management lifecycle. High-integrity closeout documentation provides the lifeblood for four core reliability functions:

1. Reliability Engineering & Bad-Actor Elimination: Reliability engineers cannot execute Weibull life-data modeling, Crow-AMSAA trend analysis, or Root Cause Failure Analysis (RCFA) without knowing the exact component that failed and the physical degradation mechanism.
2. Job Plan Calibration & Continuous Improvement: Maintenance planners cannot refine task steps or improve labor estimates if actual work hours and parts discrepancies are never reported back.
3. MRO Storeroom Inventory Accuracy: Automated inventory replenishment, min/max calculations, and stock reservation systems depend on accurate records of parts consumed during execution.
4. Regulatory & Process Safety Compliance: Under regulatory frameworks (e.g., OSHA Process Safety Management [PSM] 29 CFR 1910.119, EPA Risk Management Plan [RMP], and ISO 55001), documenting testing results, relief valve lift pressures, and pressure vessel wall thickness readings is a non-negotiable legal requirement.

---

Proportional Technician Field Feedback Requirements

Define required closeout fields by work type, risk, and analytical need. The following five categories are commonly useful; not every category applies to every work order:

[1. Actual Labor Hours]        ──> (By craft, individual technician, standard vs. overtime)
[2. Actual Parts Consumed]     ──> (Exact part/lot numbers, quantities used vs. returned)
[3. As-Found / As-Left State]  ──> (Quantitative measurements: clearances, runout, torque, megger)
[4. Observable Symptoms]       ──> (Operating context: acoustic, thermal, fluid leakage, alarms)
[5. Action Taken & Mechanism]  ──> (Restorative steps executed & physical physics of failure)

1. Actual Labor Hours Worked

Technicians must log the exact elapsed hours spent on the job, broken down by individual craftsperson and labor rate category (standard time vs. overtime). This allows planners to calculate planning estimate accuracy:

$\text{Estimate Accuracy (\%)} = \left( 1 - \frac{|\text{Actual Hours} - \text{Estimated Hours}|}{\text{Estimated Hours}} \right) \times 100\%$

2. Actual Materials and Spare Parts Consumed

Technicians must record every item drawn from storeroom kits, pulled from truck stock, or purchased via emergency purchase orders, including manufacturer part numbers, lot numbers, and exact quantities. Crucially, any unused kitted components must be explicitly recorded and credited back to storeroom inventory rather than hoarded in personal toolboxes or craft benches.

3. Quantitative "As-Found" vs. "As-Left" Equipment Conditions

Subjective descriptions such as "pump adjusted" or "fan aligned" are unacceptable. Technicians must document precise, verifiable engineering measurements:

• Mechanical Assets: Dial indicator shaft runout ($< 0.001$ in), laser alignment angularity and offset values, soft-foot readings, dynamic vibration velocity (in/s RMS), bearing radial internal clearances, and final bolt torque values ($ft\cdot lb$).
• Electrical Assets: Insulation resistance (Megger values in $M\Omega$), winding resistance balance across phases, phase-to-phase operating voltages, running full-load amperage (FLA), and infrared thermal delta ($^\circ C$).
• Instrumentation & Valves: Bench set pressure, stroke travel time (seconds), seat leakage class verification, and 4–20 mA signal calibration check values ($0%, 25%, 50%, 75%, 100%$).

4. Observable Failure Symptoms & Operating Context

Documentation of how the failure manifested prior to shutdown: abnormal acoustic signatures (screeching, grinding), fluid leak rates, process fluid temperature/pressure spikes, or automated control system alarms.

5. Specific Corrective Actions Executed

A concise technical summary detailing exactly what was disassembled, cleaned, machined, repaired, or replaced, confirming that post-maintenance testing validated functional restoration.

---

Standardized Failure Coding Hierarchies: The ISO 14224 Standard

A universal challenge in CMMS data governance is the "free-form text trap." When technicians describe failures exclusively in unstructured narrative fields, no two work orders read the same. One technician writes "seal blew out", another writes "stuffing box leaking", and a third writes "bad packing." Because text strings cannot be aggregated, automated software cannot generate Pareto charts or identify systemic failure trends across thousands of work orders.

A site can use controlled failure fields informed by ISO 14224 (Petroleum, petrochemical and natural gas industries - Collection and exchange of reliability and maintenance data for equipment). The following four-field workflow links the affected item with mode, mechanism or cause, and remedy; it is a practical data design, not a claim that every work order must populate four mandatory levels:

[Level 1: Maintainable Item / Object]
     │  (The specific physical component that degraded — e.g., Mechanical Seal)
     ▼
[Level 2: Damage / Failure Mode]
     │  (The observable functional manifestation — e.g., External Fluid Leakage)
     ▼
[Level 3: Failure Mechanism / Cause]
     │  (The fundamental physics of failure — e.g., Abrasive Particulate Wear)
     ▼
[Level 4: Remedy / Corrective Action]
        (The restorative maintenance task — e.g., Replaced with New OEM Seal)

ISO 14224 Failure Coding Structure with Industrial Examples

The following table illustrates the application of the ISO 14224 failure coding taxonomy across three common industrial asset classes:

| Asset Class & Tag | Maintainable Item (Object) | Failure Mode (Observable) | Failure Mechanism / Cause (Root Reason) | Remedy (Corrective Action) | Reliability Engineering Follow-Up Action | | :--- | :--- | :--- | :--- | :--- | : | | Centrifugal Slurry Pump (P-101A) | Mechanical Seal Face | Excessive external process fluid leakage | Abrasive particulate scoring due to flush water seal quench failure | Replaced seal assembly; flushed and cleared barrier fluid piping | Upgrade flush plan piping from API Plan 11 to Plan 32 with external clean flush | | Centrifugal Slurry Pump (P-101A) | Radial Drive-End Bearing | High high-frequency vibration & elevated temperature | Inadequate lubrication / grease starvation leading to severe spalling | Replaced bearing; dynamic laser aligned to $< 0.002$ in | Recalibrate automated grease dispenser interval; audit craft lubrication route | | AC Induction Motor (460V) (M-204B) | Stator Winding | Electrical trip on earth fault / ground overcurrent | Thermal insulation breakdown from continuous operational overloading | Rewound stator coils; cleaned and re-dipped insulation varnish | Install RTD temperature monitoring; partner with Operations to lower line throughput | | Pneumatic Control Valve (FV-302) | Actuator Diaphragm | Valve fails to stroke / sluggish response | Elastomer fatigue cracking due to ozone exposure and mechanical age | Replaced elastomer diaphragm; bench calibrated actuator spring | Update master PM schedule to replace elastomer diaphragms every 36 months | | Industrial Gearbox (GB-501) | High-Speed Pinion Gear | Abnormal gear meshing acoustic chatter | Pitting and micro-spalling from dynamic angular shaft misalignment | Replaced pinion gear set; realigned input shaft with laser system | Mandate precision laser alignment and dynamic thermal growth compensation |

---

Eliminating Generic Failure Codes: "The Trap of Generic Codes"

Generic catch-all codes reduce analytical value, but deleting them without a usable alternative can encourage false precision. Limit and govern entries such as:

• Broke
• Worn Out
• Normal Wear and Tear
• Repaired / Fixed
• Routine Maintenance
• Operator Error / Other

The Behavioral Trap of Lazy Coding

In human factors engineering, workers under operational time pressure naturally follow the path of least cognitive resistance. If a coding list is hard to use or forces a cause before evidence exists, workers may overuse generic selections. The resulting CMMS data then becomes difficult to analyze, rendering multi-million-dollar software investments incapable of answering basic reliability questions.

CMMS Configuration Best Practices

To enforce high-integrity data entry, reliability managers implement specific system constraints:

• Cascading Logic: Technicians cannot select a Failure Cause until they have first selected the specific Maintainable Item and Failure Mode. For example, selecting "Bearing" limits subsequent options to bearing-specific mechanisms (fluting, spalling, brinelling, starvation).
• Proportional required fields: Require the fields needed for that work type. Do not force all four failure fields when a cause is unknown or the work is not a failure event; allow an analysis status and accountable later update.
• Governed unknown/other states: Allow a controlled unknown or other selection when evidence is insufficient, require useful context, and route recurring uncoded events for taxonomy review. Do not force a technician to invent a cause.

---

Closing the Technician-to-Planner Feedback Loop

Maintenance planning is not an isolated, static event; it is an iterative, closed-loop learning system. A standard job plan created in the planning office is merely an engineering hypothesis. Only physical execution by craft technicians in the field tests that hypothesis against reality.

[Planner Creates Job Package]
       │  (Estimates labor hours, defines steps, specifies parts BOM)
       ▼
[Craft Technician Executes Job in Field]
       │  (Discovers unexpected piping strain, incorrect bolt length, tight clearances)
       ▼
[Technician Redlines Work Order at Closeout]
       │  (Documents actual hours, correct part numbers, missing specialty tools)
       ▼
[Supervisor Audits & Endorses Feedback]
       │  (Reviews completeness and routes to Planning Department)
       ▼
[Planner Updates Master Job Plan & BOM in CMMS]
       │  (Future work packages automatically deploy with flawless precision)
       └──────> (Continuous Improvement Loop)

Three Critical Feedback Loops

1. Equipment Bill of Materials (BOM) Refinement: When a mechanic disassembles a pump and discovers that an unlisted gasket or non-standard snap ring was required, documenting that component allows the planner to add it permanently to the asset's CMMS BOM. The next time that pump is planned, the kit will contain the exact parts, preventing a two-hour delay.
2. Labor Estimate & Skill Calibration: If a job plan estimates 2 hours for a motor replacement, but experienced electricians consistently require 4 hours due to complex conduit geometry, the planner must calibrate the master estimate. Failing to update estimates results in chronically over-scheduled shifts and broken schedule compliance.
3. Task Procedure & Tooling Upgrades: When technicians identify that a specialty offset wrench or hydraulic puller is necessary to remove a seized coupling safely, the planner updates the standard job plan to mandate staging that tool in the future.

Exam Tip: Establishing this feedback loop requires high psychological safety and active leadership reinforcement. When planners review technician redlines, update the job plan, and personally thank the technician, craft engagement surges. Conversely, if technicians write detailed feedback on paper work orders that disappear into a supervisory file cabinet, craftspeople rapidly stop providing notes.

---

Work Order Technical Completeness Audits

In SMRP work management architecture, there is a fundamental distinction between Technical Completion (TECO) and Business / Financial Closeout:

• Technical Completion (TECO): Signifies that physical work is finished, the asset has been tested and safely handed back to Operations, and the technician has logged all actual hours, materials, measurements, and failure codes. TECO is governed by Maintenance Supervision.
• Financial / Business Closeout: Signifies that all accounting variances have been reconciled, external contractor invoices cleared, and costs transferred to the general ledger. Governed by Accounting/Finance.

The Frontline Supervisor as Quality Gatekeeper

Assign a clear quality gate for technical completeness. The responsible reviewer should check closeout within a locally defined time based on risk and workflow, and should investigate missing or implausible evidence instead of merely batch-approving records.

A reliability department can conduct periodic Work Order Technical Completeness Audits using a risk-based or statistically designed sample. The cadence, sample size, weights, and thresholds below are illustrative and must be validated locally:

Post-Work Documentation Audit Scorecard