Preventive vs Predictive Maintenance for Planned Shutdowns
Field brief 01
Signal → Decision → Shutdown
Written by Miya Zheng
Sales Director
Technical review Moore Automated
Product Verification Team
Published 10 September 2026
Last reviewed 10 September 2026
For maintenance engineers, plant managers and MRO procurement
Choose the maintenance method before the shutdown chooses it for you
Preventive maintenance protects known service intervals. Predictive maintenance uses condition data to support better timing. Most plants need a deliberate mix of both—not a promise that every failure can be predicted.
The practical question is not which method sounds more advanced. It is which method fits each asset’s criticality, failure behaviour, available data, safety requirements and spare-parts risk. A useful strategy connects condition signals to engineering decisions, approved work scopes and materials that can arrive before the shutdown window opens.
One operating principle
Monitor what matters, define who acts on the evidence, and confirm the exact automation parts before the planned outage begins.
01
Estimate downtime exposure without relying on a generic headline number
Downtime exposure per event
A site-specific planning framework—not an accounting rule. Use plant finance, operations and safety data to assign the figures.
Lost production value Contribution margin or service capacity unavailable during the event.
Idle and recovery labour Operators, maintenance staff, contractors and overtime.
Restart and quality loss Scrap, rework, cleaning, validation and controlled ramp-up.
Response and supply cost Diagnosis, repair, replacement parts and expedited freight.
Customer impact Late delivery, lost capacity and agreed service consequences.
Safety and compliance impact Evaluate with the site’s qualified safety and compliance teams.
02
Preventive, condition-based and predictive maintenance compared
Approach
Primary trigger
Best fit
Main limitation to manage
Reactive maintenance
Failure has already occurred
Low-criticality assets where failure creates no unacceptable safety, environmental or production risk
Unplanned timing, secondary damage and emergency sourcing
Preventive maintenance
Calendar time, operating hours, cycles or a prescribed interval
Assets with meaningful service intervals, statutory checks or known age-related tasks
Work may be performed too early or may not detect an unrelated developing fault
Condition-based maintenance
A measured condition crosses a defined inspection or action threshold
Assets where vibration, temperature, oil, electrical or process measurements reflect deterioration
Poor baselines, sensor placement or alarm governance can create misleading signals
Predictive maintenance
Analysed trends or models indicate a developing failure and an estimated intervention window
Critical assets with dependable data, repeatable failure signatures and qualified analysis
Predictions are probabilistic and depend on data quality, context and model maintenance
03
Select the strategy asset by asset
QUESTION 01
How critical is failure?
Rank safety, environmental, production, quality and customer consequences. High criticality justifies stronger controls and earlier spare-parts decisions.
QUESTION 02
Can deterioration be observed?
Confirm that a measurable parameter changes early enough to support action. Not every failure mode provides a useful warning.
QUESTION 03
Is the data trustworthy?
Review sensor location, calibration, sample rate, baseline, operating state and ownership of alarm review.
QUESTION 04
Can the plant act in time?
A warning is valuable only when engineering approval, labour, permits, tools and verified parts can be ready before intervention.
From signal to action
Condition monitoring is a decision system, not only a sensor system
Define the evidence path before collecting more data. Each alert should have an owner, a validation step, an escalation rule and a link to the maintenance and material plan.
1. Establish the baseline Record normal operating states, loads, speeds and environmental conditions.
2. Define alert logic Set inspection and action thresholds with qualified reliability specialists.
3. Validate the signal Check instrument health, process context and corroborating measurements.
4. Translate risk into work Create an approved scope, safety plan, labour estimate and shutdown duration.
5. Confirm material readiness Verify full part numbers, revisions, condition, evidence and delivery timing.
6. Review the outcome Compare the prediction with inspection findings and update the baseline.
04
Build a shutdown-ready automation spares plan
Freeze the installed baseline Capture cabinet drawings, complete product labels, hardware revisions, firmware, software and network dependencies.
Separate exact spares from substitutes An available successor may still require engineering, wiring, logic, communication or certification changes.
Request current evidence Ask for photographs of the offered unit, label, connectors, packaging and quantity—not only catalogue images.
Agree the inspection scope State which identity, visual, electrical or functional checks are possible and which checks remain for site acceptance.
Plan the logistics backwards Allow time for review, payment, export documents, freight, customs and an on-site receiving inspection.
Protect the restart Prepare backups, rollback steps, configuration control and post-maintenance verification before the equipment is isolated.
05
Supplementary video
Video example: verify proximity-system test equipment
This Moore Automated video presents a Bently Nevada TK3-2E Proximity System Test Kit. It provides a practical reference for reviewing the supplied equipment and visible product details before a planned maintenance or shutdown task; it is not site-specific testing or installation instruction.
Related products
Condition-monitoring and machinery-protection components
Use the product routes below to review representative components. Send the complete installed part number and revision to Moore Automated before treating any item as an exact replacement.
BENTLY NEVADA 3500/22M 138607-01 Transient data interface BENTLY NEVADA 3500/42E 285691-01 Proximitor and seismic monitor VIBRO-METER VM600 CMC16 200-530-025-014 Condition monitoring card EPRO PR6423/003-030 + CON021 Eddy-current signal chain
06
Related customer success cases
PROJECT READINESS
How an Indonesian Partner Verified Bently Nevada 3500 Hardware Before Project Deployment
Read the machinery-protection case ON-SITE VERIFICATION
How a Middle Eastern Client Verified Siemens Parts On Site in China
Read the inspection case INTERNATIONAL DELIVERY
Sudan Customer Verifies ABB Hardware On Site Before International Delivery
Read the supply case
Editorial transparency
Information sources
These sources support the maintenance definitions, condition-monitoring framework and shutdown-safety context used in this article.
U.S. Department of Energy — Operations and Maintenance Best Practices Operations, maintenance and efficiency guidance for facility practitioners.
ISO 17359:2018 — Condition monitoring and diagnostics of machines General procedures for establishing a machine condition-monitoring programme.
OSHA 29 CFR 1910.147 — The control of hazardous energy U.S. regulatory context for controlling hazardous energy during servicing and maintenance. Local requirements may differ.
IBM — Preventive maintenance vs. predictive maintenance Background comparison of preventive, condition-based and predictive approaches.
Manufacturer documentation for the exact installed equipment Use current manuals, service instructions, product notices and project records for the specific model and revision.
07
Frequently asked questions
What is the main difference between preventive and predictive maintenance?
Preventive maintenance is normally triggered by time, use or a prescribed interval. Predictive maintenance analyses condition trends to estimate when intervention may be needed. The correct choice depends on asset criticality, failure behaviour, available data and the plant’s ability to act.
Does predictive maintenance eliminate unplanned downtime?
No. It can improve visibility for failure modes that produce measurable warning signs, but it cannot predict every event. Sensor quality, baselines, operating context, analysis and response discipline all affect the result.
Which condition-monitoring methods are commonly used?
Depending on the asset and failure mode, programmes may use vibration, temperature, oil or lubricant analysis, electrical measurements, ultrasound and process-performance data. Qualified specialists should select and interpret the method.
How should a plant estimate the cost of downtime?
Use site-specific figures for lost production value, idle and recovery labour, scrap or rework, diagnosis and repair, expedited supply, restart losses and customer impact. Safety and compliance consequences require separate qualified review.
When should MRO procurement become involved?
Involve procurement when the asset-criticality review identifies long-lead, scarce, obsolete or revision-sensitive components. Early involvement creates time to verify identity, condition, documentation, inspection scope and logistics before the shutdown.
What information should I send for a shutdown spare-parts review?
Send the manufacturer, complete part number and suffix, hardware revision, firmware, quantity, required condition, installed-system details, clear product-label photographs, destination and required on-site date.
Article-specific inquiry
Build a shutdown-ready spare-parts brief
Moore Automated can review availability and sourcing evidence for current, scarce and discontinued industrial automation parts. Share the installed identity and required date so the team can distinguish an exact match from a proposed alternative.
Request a shutdown review Email the part list
Include these six items
Manufacturer and complete order code
Hardware revision and firmware, if known
Installed system and asset criticality
Quantity and required condition
Clear label and connector photographs
Destination and required on-site date
Technical and safety note: Moore Automated is an independent industrial automation parts supplier. This article provides planning and procurement guidance, not site-specific engineering or safety instructions. Maintenance, compatibility, hazardous-energy control and restart approval remain the responsibility of the customer’s qualified personnel under applicable procedures and regulations. Product names and trademarks belong to their respective owners.