Post 05 of 12
Choose a maintenance strategy by failure mode, data availability and operational consequence.
Preventive maintenance follows a defined interval or usage threshold. Condition-based maintenance responds to observed equipment condition; predictive methods can help assess how that condition may develop. A fleet can use these approaches together. The right choice starts with the asset and the required decision, not a promised savings percentage.
Retain prescribed inspections and service intervals where the equipment instructions, approved maintenance plan or applicable requirements call for them. Predictive monitoring does not automatically justify extending an interval.
Use preventive tasks where degradation is linked to usage or age and monitoring would not improve the decision.
Keep task history, inspection findings and parts records: they can improve the baseline for later condition monitoring.
Monitoring is useful where a detectable signal develops before a consequential failure and there is time to respond. The equipment and operating environment determine suitable measurements.
Check that sensors are reliable and that historical readings can be compared under similar operating conditions.
Assign responsibility for alert review and define the evidence needed before opening or approving a work order.
Most marine maintenance is still preventive, and for good reason. A planned maintenance system schedules tasks by calendar or running hours, records what was done and keeps the evidence that surveyors and auditors expect. The ISM Code requires companies to maintain ships and equipment in line with the relevant rules and their own procedures.
Tasks and intervals taken from maker's instructions, class requirements and company experience.
Running-hour counters kept accurate, because many machinery intervals depend on them.
Critical equipment identified, with tests and inspections for systems whose sudden failure could cause a hazardous situation.
Findings recorded in a usable form: what was measured, what was replaced and what condition parts were in.
Spare parts linked to tasks, so the parts are on board when the job is due.
The choice is made per equipment type and failure mode, not for the whole vessel. A typical pattern looks like this, subject to maker's guidance and class:
Main and auxiliary engines: preventive overhauls by running hours stay in place; monitoring of exhaust temperatures, cylinder pressures and oil analysis helps target inspections between them.
Pumps, fans, compressors and electric motors: strong candidates for condition monitoring, because vibration and temperature usually change well before failure.
Separators and purifiers: routine cleaning stays preventive; vibration and throughput trends can show developing problems earlier.
Steering gear: mandatory tests and inspections remain preventive; monitoring can add early warning on hydraulic pumps.
Lifesaving and firefighting equipment: inspection and testing intervals are prescribed and remain preventive.
Hull and coatings: periodic inspection and survey; performance data can indicate fouling between dry-dockings.
Illustrative example: scheduled servicing remains in place for a component with a prescribed interval, while a critical pump is monitored for a developing vibration trend. Engineers review the trend and decide whether an additional inspection is needed. These are complementary decisions, not evidence that one strategy is always cheaper.
Build a comparison from maintenance hours, parts, monitoring costs, review effort and downtime. Distinguish observed costs from estimated avoided losses.
Use comparable time periods and explain changes in operating hours, fleet composition and workload.
Include false alerts, sensor maintenance and integration effort in the monitoring cost.
Report measured results and assumptions separately; retain scheduled obligations unless an approved change permits otherwise.
No. Monitoring adds sensors, data work and review effort. It pays off on equipment where failures are costly and detectable early. For simple, low-cost items, a preventive interval is often the better choice.
Only where maker's guidance, company procedures and class allow it, and usually through an approved condition-based arrangement. Monitoring evidence supports the case; it does not change obligations by itself.
With one or two critical machines that have a known failure mode, existing data and a clear owner. The predictive maintenance guide lists the data to check and how to run a pilot.
Next steps
Choose a pilot where the failure mode, measurable signal and accountable response are clear. Use the ship maintenance guide for the data checklist and the monitoring example to explore the workflow.
How Delta BI can help
Predictive maintenance software for ships and industrial assets: SmartMaintain analyses sensor data to flag developing failures and prioritise inspections.
See how it worksStart with the decision
Tell us about your challenge. We will reply to arrange an assessment and identify a practical route to delivery.
Request a data assessment