A failed photoelectric sensor stopped a high-speed packaging line twice in one month. The replacement itself took less than 20 minutes. Finding the correct model, confirming wiring and sensing range, and waiting for a part took far longer. This machine downtime reduction case study examines how a plant addressed that pattern by treating repeat stoppages as a maintenance and procurement problem, not just a repair event.

The example is a representative composite based on common issues in packaging and material-handling operations. The numbers are illustrative, but the approach applies to plants running PLC-controlled conveyors, fillers, cartoners, palletizers, and similar equipment. The central lesson is straightforward: downtime falls when teams identify the failure mode, protect critical parts availability, and verify the surrounding control system before restarting.

The operating problem

The plant operated two packaging lines across three shifts. Line 2 had become the main source of unplanned downtime, averaging 18.5 hours per month over a six-month period. Most events were classified simply as electrical or sensor faults in the maintenance log. That label was enough to close a work order, but not enough to prevent the next stop.

Production supervisors saw the line as unreliable. Maintenance technicians saw a mix of unrelated minor failures. Procurement saw urgent, irregular orders for controls components. Each group had part of the picture, but nobody had a usable record tying a specific machine symptom to a part number, root cause, repair time, and stock position.

The line was not failing because every component was old. It was failing because a handful of recurring issues created long recovery times. A sensor bracket would vibrate loose, an unshielded cable would develop intermittent noise near a VFD, and a damaged power supply terminal would cause a PLC input group to drop out. None of these conditions was especially difficult to repair. The delay came from diagnosis and parts identification.

Machine downtime reduction case study: finding the real losses

The maintenance manager started by reviewing every Line 2 stop longer than 15 minutes. Rather than relying on broad fault categories, the team recorded the asset location, fault code, failed or suspected component, exact replacement part number, time to diagnose, time to obtain the part, and time to restart.

After eight weeks, four sources accounted for most lost production time:

• Photoelectric sensors and damaged sensor cables at accumulation zones
• A 24 VDC power supply operating near its load limit
• Intermittent input signals at a remote I/O block near a VFD
• Delays caused by missing or incorrectly specified replacement parts

The distinction between failure frequency and downtime impact changed the priority list. A sensor issue occurred often, but normally caused a short stop when a correct spare was available. The power supply had failed only once, yet the resulting outage lasted nearly six hours because the plant did not have the exact unit or a verified compatible replacement on site.

That is a common maintenance planning problem. The most frequently replaced item is not always the most critical spare. A plant needs to consider failure rate, lead time, replacement complexity, and the production consequence of a stockout.

Separating the component failure from the system condition

The team also stopped replacing components without checking why they failed. On one accumulation conveyor, three sensors had been replaced in five months. The original assumption was poor sensor reliability. Inspection found that the mounting bracket allowed gradual movement from conveyor vibration. The sensor face then shifted outside its reliable detection range.

The corrective action was not another sensor order alone. Maintenance installed a more secure bracket, confirmed the sensing distance against the product variation, and added the sensor alignment point to the weekly inspection route. A spare sensor was still required, but the underlying cause was mechanical.

At the remote I/O station, the issue was electrical. A signal cable ran too close to a VFD output cable after a previous field modification. The intermittent fault did not always appear during troubleshooting, which led technicians to suspect the PLC and I/O hardware. Cable routing and shielding corrections resolved the signal noise. Replacing the I/O block would have been expensive and would not have solved the problem.

This is why downtime reduction cannot be reduced to stocking more parts. Inventory protects recovery time. Root-cause work protects recurrence. Both are necessary.

Building a critical-spares plan around exact part numbers

The plant's previous storeroom process used generic descriptions such as "proximity sensor" or "PLC power supply." That caused avoidable confusion when a technician needed a particular output type, connector style, voltage rating, or communication interface. A close-looking part may not be electrically compatible, mechanically suitable, or accepted by the machine program.

The new critical-spares list used manufacturer, exact part number, machine location, function, quantity on hand, reorder point, and approved alternates where applicable. For controls equipment, the team also recorded firmware, configuration files, terminal blocks, memory cards, connectors, and any required programming steps. A spare PLC CPU without a current program backup is not a complete recovery plan.

For the Line 2 equipment, the initial list included sensors, cables, 24 VDC power supplies, relays, contactors, safety components, remote I/O modules, HMI accessories, and selected PLC hardware. Not every item was stocked in duplicate. The plant applied different rules based on consequence.

Low-cost, high-use sensors and connectors were held in quantities that could cover expected demand. Higher-value modules were stocked only when they had long replacement lead times, limited interchangeability, or the ability to stop the full line. Some parts were not held on site, but their exact model numbers, approved sourcing path, and required delivery window were documented in advance.

For multi-brand equipment, centralized sourcing can reduce the time spent locating an exact replacement. Industrial buyers should still verify the full manufacturer part number, electrical specifications, revisions, and machine compatibility before ordering. American Automation 24 supports this type of part-number-driven purchasing across major automation brands, which is useful when a repair requires components from more than one manufacturer.

Avoiding the wrong-spare problem

A spare is only useful if it can be installed safely and quickly. The team found two items in the cabinet that were labeled as replacements but had different pinouts from the installed devices. Another spare was from an obsolete series and required a different configuration tool.

To correct this, maintenance tested selected spares during scheduled downtime. Technicians checked connector fit, terminal layout, program transfer requirements, I/O addressing, and expected startup behavior. They also photographed cabinet locations and attached the verified part number to the asset record.

This work adds effort upfront. It is usually cheaper than discovering incompatibility in the middle of a production stop.

Standardizing diagnosis and restart work

The plant created short fault-response instructions for the recurring failures. These were not long manuals. Each instruction began with the machine symptom and listed the first checks in the order most likely to isolate the issue.

For example, a missing product detection signal required the technician to check sensor indicator status, target alignment, supply voltage, connector condition, and PLC input status before replacing the sensor. A 24 VDC fault required a load check, terminal inspection, output voltage measurement, and review of recently added devices. This sequence prevented technicians from replacing good parts based on a fault message alone.

The plant also improved escalation rules. If a stop exceeded 30 minutes, the shift lead notified maintenance, controls engineering, and the parts coordinator at the same time. That prevented a common delay where procurement learned about an urgent need only after several hours of unsuccessful troubleshooting.

Controls backups were placed under formal revision control. The team saved PLC and HMI files after approved changes, recorded the software version, and tested restore procedures on non-production equipment where possible. This did not eliminate hardware failures, but it reduced risk when replacing a controller, HMI, or communication module.

Results after six months

Six months after the changes began, Line 2's average unplanned downtime declined from 18.5 hours to 9.2 hours per month. The line did not become failure-free. Sensors still required occasional replacement, and mechanical wear still created maintenance work. The difference was that recurring faults were corrected more completely and replacement parts were available with less uncertainty.

Mean time to repair for the most common sensor-related stops dropped from 74 minutes to 29 minutes. The power supply issue did not recur after load adjustments and terminal repairs, but the plant added an approved replacement unit to its critical-spares strategy. Emergency purchases declined because technicians and buyers no longer had to identify basic replacement requirements during an outage.

The trade-off was additional discipline. The team spent time cleaning asset records, verifying part numbers, checking cabinet conditions, and maintaining stock data. It also tied up some capital in spares. For a line with meaningful production value, that cost was justified. For a noncritical asset with readily available parts, a smaller stock position may be the better decision.

Applying the approach to your equipment

Start with the stoppages that consume the most time, not simply the ones that happen most often. For each event, capture the failed condition, the exact component involved, the elapsed time spent diagnosing it, and whether the correct replacement was immediately available. That record will quickly show whether your biggest opportunity is better preventive work, a control-system correction, verified spares, or faster purchasing.

The useful next step is not to fill a storeroom indiscriminately. It is to make sure the parts that can stop a critical machine are identified by exact model number, checked for compatibility, and connected to a clear response plan before the next failure puts production on hold.