Mean Time to Repair: Commissioning and Acceptance Checklist

Mean time to repair (MTTR) is normally treated as a maintenance performance metric, but in practice it is largely determined before a system ever enters production. Commissioning and acceptance testing are the moments when physical access, cabling, labelling, diagnostics, fault records, and spare-part decisions become committed. Once production begins, those choices either compress or extend every future repair. This article explains how warehouse operators, maintenance engineers, and controls teams can use a commissioning and acceptance checklist to shape MTTR deliberately rather than inheriting it as an afterthought.

Mean Time to Repair Begins at Commissioning #

MTTR is the total time from the moment a fault is noticed to the moment the equipment is returned to a normal operating state. It includes detection, diagnosis, energy isolation, physical access, part replacement, re-commissioning, and verification. It is not the same as the time spent with a wrench in hand. A repair can be completed in ten minutes, yet still contribute two hours to MTTR because the technician could not reach the component, the part was not stored on site, or the fault code did not point to the correct subsystem.

Commissioning is the last stage of the project where such constraints can be corrected without shutting down a live operation. This is why MTTR is best understood as a design property. The layout of a conveyor zone, the position of a READY sensor, the type of connector used on a variable frequency drive, and the strictness of the failure-coding schema are all decided somewhere between the contractor’s installation and the final acceptance sign-off. An acceptance checklist that ignores repair time will certify a system that runs well but cannot be maintained efficiently.

Operating Context and Component Interactions #

Warehouse material handling systems are rarely single-machine problems. A typical automated area combines conveyors, sorters, palletizers, stretch wrappers, transfers, elevators, and an automated storage and retrieval system, all coordinated by programmable logic controllers, safety relays, network switches, and drives. These subsystems interact in ways that complicate diagnosis. A sorter fault can cause accumulation on upstream conveyors; a failed network switch can make several PLC racks appear unresponsive; a blocked safety interlock can freeze a single zone while the rest of the line continues to feed material into it.

Observable symptoms therefore need to be interpreted against system interaction, not component location. A “stalled” conveyor may actually have a healthy motor and drive, with the real fault being a missing confirmation signal from a downstream photoeye. A palletizer that repeats a cycle may be responding to a worn limit sensor, a loose chain, or a PLC program state change. During commissioning, the maintenance team should record not just that a device works, but how its failure presents in the controls system. This forms the basis of future diagnosis and shortens MTTR from the first breakdown onward.

Commissioning Evidence That Predicts Repair Time #

Acceptance evidence should predict repairability, not only functionality. The table below lists the types of evidence normally gathered during commissioning and the repair-time risk that remains if that evidence is not collected.

Each row in the table translates a commissioning activity into a measurable MTTR influence. The commissioning team should treat these as acceptance criteria, not merely as project paperwork.

Access, Isolation, and the Physical Constraint #

Physical reach is often the largest and most predictable component of MTTR. A sensor mounted behind a drive roller, a junction box placed under a transfer deck, or a network switch located above an aisle of tall racks all force the technician into slow and unsafe working positions. The time required to remove guarding, position a platform, apply lockout, and verify zero energy is part of MTTR and should be tested during commissioning.

Every energy source must be considered. Conveyor systems may have electrical, pneumatic, and gravitational energy; automated storage and retrieval machines may have powered masts and moving carriages. The acceptance checklist should confirm that isolation points are accessible, clearly identified, and capable of being locked out independently. Site lockout procedures, OEM documentation, and competent engineering judgment take priority over any commissioning shortcut, and no instruction in this article overrides those requirements. A design that cannot be safely isolated within a reasonable time should be challenged before acceptance, not discovered during the first emergency repair.

Decision boundary: if a high-rate failure component cannot be reached, isolated, and replaced under supervised commissioning conditions, the design has already failed a repair-time requirement. Correcting it after handover is more expensive and more disruptive.

Failure Coding and Condition Evidence #

MTTR cannot be reduced with data that cannot be interpreted. Failure coding is a commissioning deliverable. It should be designed before the system is live so that every technician records the same meaning in the same fields. A failure code for “conveyor motor” is far too broad. Useful codes capture the breakdown mode, such as “motor—insulation failure,” “motor—bearing seizure,” “motor—thermal trip,” and “motor—no supply present.” This separation allows the maintenance team to see repeat patterns and address them at root-cause level rather than replacing the same part repeatedly.

Condition evidence should also be gathered during commissioning. Photographs of normal belt tracking, baseline vibration readings for rotating equipment, thermal images of drive cabinets, tension settings, and alignment values provide a reference for later comparisons. When a fault occurs, the technician can compare current readings to the accepted baseline and decide whether a slow degradation has been overlooked. This is particularly important for repeat-fault reduction, where a single recurring code without accompanying condition evidence is almost impossible to interpret correctly.

Common interpretation error: coding the replaced component rather than the failure mechanism. If a sorter diverts late because a photoeye window is dirty, coding the repair as “photoeye replaced” hides the real issue. The correct code records the failure mode and the contributing condition so that the hygiene or airflow problem can be corrected.

Spares, Diagnostics, and the First-Swap Exercise #

Spares positioning is a commissioning decision. The acceptance process should define which items are held locally, which are held centrally, and which are documented as call-off items. Criticality groups should be based on failure consequence and repair frequency. A common warehouse approach is to keep a small local stock of high-rate consumables such as photoeyes, rollers, sensors, fuses, and drive fans, while larger drives and motors are staged through a central store. The key is to make this grouping explicit during commissioning so that parts are located where they save the most time.

The most practical MTTR tool available during commissioning is the first-swap exercise. Select the ten components most likely to fail based on operational exposure and historically known failure rates, then physically remove and replace one of them under timed, supervised conditions using the same tools, access equipment, and isolation procedures that would be used during a real repair. Record the time from isolation to return-to-service. This provides a realistic MTTR baseline and exposes problems that no written review will catch. A connector that requires a special tool, a sensor that can only be removed after unbolting a guard rail, or a motor that needs a different lifting sling will become obvious during the exercise.

Diagnostic quality matters as much as spares. The controls team should verify that every alarm gives a meaningful message, that fault codes in the PLC