Condition monitoring does not begin the day after handover; it begins during commissioning. The commissioning and acceptance phase is the only period in an asset’s life when the system is new, the installation is inspectable, and the operating envelope can be deliberately exercised. A structured condition monitoring checklist used during this phase captures the baselines, verifies the detection capability of future routines, and identifies the defects that would otherwise appear later as repeat failures. This article provides a technical acceptance checklist for warehouse operators, maintenance engineers, and controls teams. It explains what to check, how to capture evidence, which interpretation errors to avoid, and how the outputs should shape the maintenance program. Site procedures, lockout requirements, OEM documentation, and competent engineering judgment always take priority over any generic checklist.
Purpose and Scope of the Acceptance Checklist #
The acceptance checklist for condition monitoring is not the same as the mechanical completion checklist or the functional test script. Mechanical completion proves that equipment was installed to drawing. Functional testing proves that a machine performs its intended sequence. The condition monitoring acceptance checklist proves that the asset can be observed, measured, and trended over its operating life. If this distinction is not made at the outset, the warehouse will inherit machines that run but cannot be effectively monitored.
The checklist should cover four deliverables. First, baseline data for all critical components. Second, clearly defined measurement points that are safe and repeatable. Third, a fault coding structure aligned with the maintenance management system. Fourth, a documented condition evidence repository that will persist beyond the commissioning team’s departure. Each deliverable must be accepted in writing by the maintenance organization, not merely by the project team.
The scope should extend to all condition-bearing assets within the material handling system: conveyor drives, sortation units, palletizers, depalletizers, automated storage and retrieval machines, hydraulic power units, air compressors, dust extraction fans, and the electrical distribution feeding them. Excluding a component because it is “new” is a common error. New equipment fails during run-in more frequently than steady-state equipment, and the commissioning period is the last chance to capture the pristine signature without production pressure.
Pre-Commissioning Conditions and Documentation Review #
Before any baseline measurement is taken, the maintenance team should verify that the installation itself meets the conditions required for meaningful monitoring. This includes checking that the equipment is correctly aligned within its specified tolerances, that foundations and anchor bolts are adequately torqued, and that flexible couplings are fitted with the correct spacer and alignment shims. A machine that is commissioned with excessive misalignment will produce baseline vibration data that is high from day one, and every future trend will be compared against a flawed reference.
Documentation review should include the OEM maintenance manual, the lubrication schedule, the electrical schematic, and the control philosophy. The condition monitoring engineer should confirm that the component designations in the manual match the asset tag numbers in the CMMS. This seems trivial, but failure coding and spare parts identification are impossible when the naming convention differs between the drawing set and the maintenance system. The review should also identify which components are non-ferrous, plastic, or belt-driven, because those assets may require different measurement techniques such as acoustic emission rather than accelerometer-based vibration analysis.
The pre-commissioning review must also confirm that access to measurement points is physically possible during operation. A vibration point located behind a guard, a thermal imaging target obscured by ducting, or an oil sampling port that requires removal of a chain guard will not be used consistently. Where access is poor, the acceptance team should document the risk and require a modification before final acceptance. If the modification cannot be made, the condition monitoring program should be designed around the accessible proxy points, and that decision must be recorded.
Operating Context and Component Interactions #
A warehouse conveyor system is not a set of independent machines. Each drive interacts with upstream and downstream equipment through the conveyed load. The condition monitoring acceptance checklist must therefore consider the operating context in which the asset will run. A sorter induction belt that is tested empty will produce a completely different vibration signature than the same belt running at full design load with cartons of varying weights. Acceptance should include baseline measurements under three operating states: no load, partial load, and full design load where safe and permissible.
Component interactions also matter across the mechanical train. An electric motor drives a gearbox through a coupling, and the gearbox drives a pulley or a sprocket. A misalignment in the coupling will load the motor bearing and the gearbox input shaft simultaneously. A worn chain sprocket will generate vibration pulses that appear in the motor’s frequency spectrum even though the motor itself is healthy. During commissioning, the team should record the baseline spectrum at each measurement point so that later analysts understand which component is the source and which is merely transmitting the vibration.
The controls interaction is equally important. A variable frequency drive changes the motor speed, which changes the bearing defect frequencies and the structural resonance points. The baseline data must be recorded at the specific speed setpoints used in normal operation, not at the nominal nameplate speed. Similarly, the connection between mechanical condition and electrical load should be understood. A conveyor that draws more current than expected during commissioning may indicate high friction from a misaligned belt or an overtensioned chain. The current signature is a condition monitoring parameter in its own right, and the commissioning baseline should include amperage readings under each operating state.
Baseline Condition Evidence Collection #
Baseline evidence is the foundation of all future trend analysis. It should be collected after the commissioning team has completed the run-in period and the machine has reached a stable operating temperature. Collecting a baseline during the first hour of operation, before bearings have run in and before lubricant has been distributed, will produce data that is not representative. The table below provides a practical guide to the baseline evidence that should be captured for common warehouse asset groups.
| Asset Group | Measurement Point | Baseline Parameter | Why It Matters |
|---|---|---|---|
| Conveyor drive motors | Drive-end and non-drive-end bearing housings | Overall vibration velocity, acceleration envelope, bearing defect frequency amplitudes, temperature | Establishes the run-in signature; later increases indicate bearing wear or lubrication failure |
| Gearboxes and reducers | Input and output bearing housings; oil level sight glass | Vibration spectrum at gear mesh frequency, oil temperature, oil sample for particle count and viscosity | Gear mesh frequency amplitude identifies gear tooth wear; oil analysis captures early debris generation |
| Fan and blower units | Bearing housings and fan casing | Vibration velocity at rotational frequency, blade pass frequency, motor current, airflow pressure | Blade pass frequency trends indicate impeller imbalance or vane buildup |
| Hydraulic power units | Pump casing and motor yoke | Vibration at pump vane/piston frequency, discharge pressure ripple, oil temperature, contamination level | Pressure ripple and vibration detect pump wear before flow loss becomes apparent |
| Automated storage and retrieval machines | Vertical lift and horizontal drive gearboxes, guide wheels | Vibration at travel speeds, position error traces, rail temperature | Early detection of rail wear or wheel misalignment prevents expensive structural damage |
| Control cabinets and power distribution | Terminal blocks, busbars, breaker connections | Infrared thermal image, temperature rise above ambient, harmonic distortion of supply | Loose connections and harmonic loads shorten component life and cause nuisance trips |
Each baseline reading should be recorded with the machine speed, load condition, ambient temperature, and the specific tool settings used. A vibration meter with a different filter setting will produce a different overall value. The evidence record must include the instrument calibration date and the exact mounting method, because a change from a magnetic mount to a stud mount changes the high-frequency response. This level of detail is often skipped, but it is precisely the detail that makes future trends reliable.
Observable Symptoms and Their Meaning #
During commissioning, the team will encounter a range of observable symptoms. The key is to record these symptoms as condition evidence, not simply to fix them and move on. A high vibration level at a fan bearing during first run may be caused by a balancing issue from shipping, a soft foot condition, or a structural resonance in the mounting frame. Each cause has a different diagnostic signature, and recording the symptom with its context allows the maintenance team to distinguish them later.
Temperature is a direct symptom of several failure modes. A gearbox that runs 20 degrees Celsius above its sibling unit on the same conveyor line indicates either an overfilled or underfilled oil level, incorrect oil viscosity, or insufficient cooling. A motor that is hot to the touch at the yoke but cool at the frame may indicate a rotor bar issue rather than an overload condition. The commissioning team should record not just the absolute temperature but the temperature difference from the ambient baseline of the facility.
Acoustic symptoms are highly informative but often underused. A high-pitched squeal from a bearing usually indicates lubrication starvation, while a low-frequency rumble may indicate a looseness issue. A repetitive clicking from a chain drive often indicates a stiff link or a worn sprocket tooth. During commissioning, much of the installation noise from construction activity is still present, so acoustic assessment should be done during quiet periods after other trades have left the area. The recorded acoustic observation should note what was heard, the location, and the operating speed at the time.
Electrical symptoms are equally relevant. Drive faults such as overcurrent or overvoltage alarms during commissioning should never be dismissed as commissioning teething issues. They are condition evidence of mechanical overload, control loop instability, or incorrect parameterization. The maintenance team should receive the complete fault log from the variable frequency drives and the PLC, with the timestamp and the operating context for each event. This log becomes the baseline for electrical condition.
Common Interpretation Errors #
The most common interpretation error in commissioning condition data is comparing the reading to a generic severity chart rather than to the machine’s own baseline. Generic charts for overall vibration velocity are useful screening tools, but a new gearbox may legitimately run at a vibration level that is flagged as “alarm” on a generic scale due to structural resonance or gear mesh frequency. The acceptance decision should be based on whether the level is acceptable to the OEM and whether it is stable, not solely on a severity chart.
Another error is averaging multiple readings across a machine. Each bearing housing is a separate measurement point with its own signature. Averaging the drive-end and non-drive-end bearing readings into a single “motor vibration” value masks which bearing is degrading. The baseline record must maintain each point individually and each trend must follow the same point location.
A third error is confusing absolute value with rate of change. A machine that runs at a moderate vibration level that increases rapidly over a week is much more concerning than a machine that runs at a higher but perfectly stable level. The commissioning baseline is the reference point for rate of change calculations. If the baseline was not recorded under repeatable conditions, the rate of change calculation is meaningless.
Interpreting lubricant analysis results requires similar caution. A single oil sample taken during commissioning can be contaminated by flushing residue from the factory test, and the particle count may be falsely elevated. The commissioning sample should be taken only after the first scheduled oil change or after adequate circulation and filtration. Even then, the first sample should be treated as a contamination check, not a wear check, because new components shed minor debris during initial seating.
Maintenance Implications and Decision Boundaries #
The condition evidence collected during commissioning directly drives the maintenance plan. Assets that show elevated baseline readings should receive a shorter initial inspection interval, even if those readings are within the OEM tolerance. Assets that show clean, stable baselines can move to a standard inspection frequency. This risk-based differentiation is the core value of a commissioning condition monitoring program. Without it, all assets receive the same generic schedule and the worst actors are discovered only after failure.
Decision boundaries must be defined before the evidence is collected. At what vibration level will the acceptance team reject a machine or require corrective work? At what temperature rise above ambient will an asset be flagged for follow-up? What rate of change over a one-week run-in period is considered acceptable? These boundaries should be documented in the acceptance protocol and agreed with the OEM where the equipment is still under warranty. Without pre-agreed boundaries, the acceptance team is in the difficult position of negotiating a rejection after the project schedule is already under pressure.
It is equally important to define what is not a rejection criterion. A minor oil leak at a fitting that can be easily sealed, a loose cable gland, or a missing label are deficiencies that should be resolved but do not affect condition monitoring acceptance. The distinction between a condition defect and a housekeeping deficiency should be explicit in the checklist, so that the maintenance team’s attention is not diverted from the data quality issues that truly matter.
If a machine fails to meet the acceptance criteria, the response should be structured. The observed condition data should be reviewed by the maintenance engineer and the OEM representative. A corrective action plan should address the root cause, not the symptom alone. For example, if the baseline vibration at a gearbox output bearing is too high, the corrective plan should investigate foundation stiffness, coupling alignment, and shaft runout before simply specifying more frequent lubrication. The re-measurement after correction also serves as a secondary baseline and becomes part of the evidence record.
Repeat-Fault Reduction and Failure Coding #
Repeat-fault reduction depends on the ability to identify a fault mode that occurs on multiple assets and to address the systemic cause. During commissioning, the condition monitoring team should establish the failure coding structure that will be used for all future work orders. The coding scheme should distinguish between the failed component, the failure mode, and the underlying cause. For example, a motor bearing failure may be coded as component “motor bearing,” mode “fatigue,” and cause “misalignment.” The same bearing failure could also be coded with cause “lubrication starvation.” The distinction matters because the corrective action is different.
The commissioning period often reveals patterns across multiple identical assets. If a fleet of twenty conveyor drives shows high motor current on all machines that are installed on the same structural bay, the cause is likely a system installation issue rather than twenty individual motor defects. These patterns must be captured in a dedicated findings register during commissioning, not buried in individual handover documents. The findings register should cross-reference the asset tag, the observed condition, the likelihood that this condition appears on other assets, and the recommended verification check.
Failure coding also feeds directly into the spare parts strategy. A commissioning finding of a failing bearing on one unit is a signal to verify whether the same bearing type is used across multiple drives and whether the spare stock is adequate. A finding of repeated oil contamination across several gearboxes is a signal to review the storage and handling of bulk lubricant in the warehouse, not to order more filters.
Spares Rationalization and Condition Evidence Retention #
Spares rationalization is often treated as a separate discipline, but it depends on condition evidence. The commissioning baseline tells the maintenance team which components are most likely to degrade first, and therefore which spares should be held. Components that show acceptable but slightly elevated baseline readings should be considered for consignment stock or a faster procurement route. Components that show textbook-clean baselines may be reviewed for potentially lower stock levels, with the understanding that no rotating asset should ever be run to failure solely due to spares strategy.
Condition evidence retention must extend beyond the commissioning handover. The evidence repository should be structured with the same asset tagging scheme used in the CMMS, and access should be granted to the reliability team, not only to the original commissioning engineers. The repository should include the raw data files, not just the summary reports. A vibration spectrum file, a thermal image, and an oil sample result all contain more information than the summary line that was written in the report. Future analysts will need the raw data when new fault signatures appear.
The retention policy should also include the configuration of the monitoring tools. If the warehouse uses a portable vibration data collector, the route points and the measurement settings should be applied to the collector as part of the commissioning handover. If the facility uses online monitoring, the software database should be populated with the asset hierarchy and the alert limits derived from the commissioning baseline. The goal is that the first scheduled condition monitoring round after handover uses the same measurement points, the same settings, and the same reference data as the commissioning round.
Final Acceptance Criteria and Handover #
Final acceptance of the condition monitoring program should be a formal milestone, documented and signed by representatives from maintenance, engineering, and operations. The acceptance criteria should include verification that all baseline measurements have been collected, all measurement points are accessible and safe, the failure coding structure is loaded into the CMMS, the raw data repository is populated, the alert limits are set, and the first scheduled monitoring round has been planned and assigned. The acceptance team should also confirm that the maintenance staff has received the training necessary to operate the monitoring instruments and interpret the reports.
The handover should include a briefing session where the commissioning condition monitoring engineer walks the maintenance team through the key findings, the anomalies, and the recommended follow-up actions. The written report is the record, but the conversation is where the context and nuance are transferred. The briefing should also cover the known limitations of the baseline data, such as measurements that were taken under partial load or at reduced speed, so that the maintenance team does not over-interpret the baseline as the complete condition picture.
Following handover, the maintenance team should schedule a deliberate review after the first 500 operating hours. This review compares the early operational condition data to the commissioning baseline and identifies any rapid changes. This point is often the first moment where a commissioning-time defect becomes visible under sustained production load. The review should be a formal checkpoint and should be used to validate the alert limits that were set during commissioning.
Key Takeaways #
- Condition monitoring acceptance is a distinct commission process, separate from mechanical completion and functional testing, and it must be formally documented and signed off.
- Baseline data must be collected under the operating conditions that the machine will actually see, including load levels, speed setpoints, and after the run-in period has concluded.
- Every measurement point must be safe to access, repeatable in its location and mounting, and recorded with the instrument settings, calibration date, and operating context.
- Compare commissioning readings against the machine’s own baseline and rate of change, not solely against generic severity charts, which can mislead on new equipment.
- The failure coding structure, spare parts strategy, and condition evidence repository must be aligned with the asset tagging scheme and populated during the commissioning phase, not after handover.
- Repeat-fault reduction starts by capturing commissioning findings in a grouped register to reveal systemic patterns across identical assets.
- Decision boundaries for acceptance or rejection must be agreed before data collection, including the limits for vibration, temperature, current, and oil condition.
- Formal handover includes training, a briefing session, and a scheduled review after the first 500 operating hours to validate alert limits and catch early degradation.
