ISO 17359 Machine Condition-Monitoring Methodology

1. What Is ISO 17359 Machine Condition-Monitoring Methodology?

The ISO 17359 Machine Condition-Monitoring Methodology is an operational reliability framework for establishing a machine condition-monitoring program. It structures the decisions connecting equipment functions, potential failures, measurements, diagnosis, maintenance action, and program improvement. The standard provides general guidance applicable to all machines rather than prescribing one technology.

ISO is the short name of the International Organization for Standardization, not an acronym with separate framework components; 17359 identifies the standard.

For manufacturers of industrial machinery, the practical question is which deterioration should be monitored, where, and how early enough to support intervention. Consultants can use its logic to structure reliability engagements involving production equipment or customers’ installed machines.

2. Origin and Background

ISO first published ISO 17359 in 2003 through its technical committee structure, specifically ISO/TC 108/SC 5, which addresses condition monitoring and diagnostics of machine systems. Its origin is collective standards development, not an individual consultant’s model.

The standard established general procedures connecting machine-monitoring activities with relevant technical standards. The current third edition, ISO 17359:2018, was published in January 2018 and confirmed in 2023. Its official title is “Condition monitoring and diagnostics of machines — General guidelines.”

Its original purpose remains relevant: organize monitoring around failure mechanisms and useful maintenance decisions. The 2018 revision also strengthened the connection to the ISO 55000 asset-management standards.

3. How ISO 17359 Machine Condition-Monitoring Methodology Works

The methodology follows a feedback loop, summarized below. It starts with business consequences and machine functions, not sensor selection.

These stages connect analysis to action and feed maintenance findings back into the program. A central decision gate is whether the relevant failure mechanism produces a measurable symptom; where it does not, alternative maintenance or redesign may be necessary.

For production equipment, measurements might include vibration on rotating assemblies, lubricant condition and wear particles in gearboxes, or thermal patterns in electrical equipment. Techniques should match the suspected fault rather than simply collect available signals.

Diagnosis concerns interpreting the machine’s present condition and likely fault. Prognosis concerns its future condition. An abnormal trend does not, by itself, establish a dependable remaining-life estimate. Diagnostic and prognostic processes require appropriate data and methods.

4. When to Use ISO 17359 Machine Condition-Monitoring Methodology

Use the methodology when selecting monitoring investments, investigating weak returns from an existing program, or standardizing reliability practices across plants. Relevant executive questions include:

• Which failure mechanisms threaten bottleneck throughput or product quality?
• Where is permanent monitoring justified over periodic inspection?
• Can a detected problem trigger action before unacceptable consequences occur?

For machine tools, distinguish equipment deterioration from changes in tooling, workpieces, and operating profiles. Manufacturing monitoring must support both downtime decisions and production-quality requirements; changing workloads complicate interpretation.

The methodology is especially powerful where consequential failures have observable precursors and sufficient intervention time. It is a weaker fit for failures without measurable warning. Monitoring intervals must reflect fault progression, operating conditions, and criticality—not an arbitrary monthly schedule.

Plan for asset records, failure histories, representative measurements, and input from production, maintenance, quality, and equipment specialists. Effort depends on inventory quality and the operating cycles needed to establish credible baselines. A scoped workshop is only the beginning of a validated program.

Treat the initiative as a maintenance and reliability improvement project, with technology serving the maintenance decision rather than defining it.

5. How to Apply ISO 17359 Machine Condition-Monitoring Methodology: Step-by-Step

The following implementation guide adapts the standard’s logic for an executive-sponsored plant program.

1. Clarify the decision, scope, and economics.

   Define the production area, equipment boundaries, planning horizon, and intended outcome. Compare installation, analysis, and response costs with avoidable downtime, quality losses, and secondary damage. Distinguish genuinely lost production from output recoverable later.

2. Gather technical and operating evidence.

   Collect asset registers, supplier manuals, work orders, failure findings, operating profiles, measurement histories, and spare-part lead times. Interview operators and technicians. Identify missing evidence explicitly rather than interpreting incomplete records as proof of reliability.

3. Define asset–failure-mode units.

   Assess specific combinations, such as a grinder spindle and bearing deterioration, rather than rating an entire machine “monitored.” Map production dependencies and standby arrangements. Keep safety consequences visible instead of burying them inside a weighted financial score.

4. Build the monitoring plan.

   Create a register linking each failure mechanism to its symptom, parameter, technique, measurement location, operating state, frequency, alert criteria, and response owner. Document excluded mechanisms and their alternative maintenance treatment. Select instrumentation only after establishing these requirements.

5. Establish baselines and challenge abnormalities.

   Collect measurements under defined, comparable conditions. Check instrument quality and machine configuration before interpreting deviations. Use additional evidence when diagnosis is uncertain; do not convert every threshold crossing directly into a replacement order.

6. Connect detection to intervention.

   Specify who validates an alert, determines urgency, authorizes work, obtains parts, and negotiates production access. Include repair verification and findings in the work-order closeout. Early warning has little operational value without an executable response.

7. Test sensitivities and scenarios.

   Challenge assumptions about deterioration speed, redundancy, spare availability, shutdown access, and demand. Test whether the conclusion changes with another product mix or operating duty. Use ranges where evidence cannot support precise forecasts.

8. Align stakeholders and improve the program.

   Pilot on a bounded asset group. Review missed faults, nuisance alerts, diagnostic confidence, response time, and verified maintenance findings. Agree which practices should be standardized and which require machine-specific judgment before expanding across plants.

Where monitoring exposes gaps in ownership and work execution, maintenance operating model design can translate the analysis into planning responsibilities, escalation paths, and consistent work-management routines.

6. Example: ISO 17359 Machine Condition-Monitoring Methodology in Action

Illustrative scenario: Aldercrest Industrial Drives, a fictional gearbox manufacturer, experiences interruptions and intermittent surface-finish problems at its final gear-grinding operation. Management considers installing identical vibration sensors throughout the plant.

The reliability team instead applies ISO 17359’s logic. It maps the production route and identifies the grinder as a bottleneck without readily available alternative capacity. Its units of analysis include spindle-bearing deterioration, hydraulic contamination, and failures of auxiliary pumps. The team uses maintenance findings, supplier guidance, operating records, quality results, and repair lead times—not generic industry benchmarks.

The resulting plan differentiates the responses. The grinder receives vibration monitoring under a repeatable operating condition; the hydraulic unit receives an oil-sampling plan; a redundant auxiliary pump retains periodic inspection. The team also identifies failure mechanisms outside the chosen monitoring coverage.

When spindle measurements change, an analyst checks that the operating profile is comparable and investigates the signal alongside temperature and quality evidence. The team recommends inspection during a planned production window, with parts and labor prepared. Inspection confirms bearing damage.

After repair, Aldercrest verifies both machine readings and product finish, records the findings, and revises its alert criteria. The decision was not to instrument everything, but to establish targeted coverage and a reliable response process.

7. Strengths and Limitations

Strengths

• Traceable decisions: The connection between equipment functions, failure mechanisms, symptoms, and intervention makes monitoring choices explainable to engineering and management.
• Focused investment: Criticality assessment helps prioritize equipment using consequences, redundancy, and monitoring economics rather than purchase price alone.
• Closed-loop learning: Maintenance findings and effectiveness reviews provide feedback for improving the program instead of leaving it as a static sensor deployment.

Limitations

ISO 17359 provides general guidance, not machine-specific diagnostic algorithms or universal alarm limits. Its usefulness depends on selecting suitable techniques and interpreting the evidence competently.

Manufacturing variability also matters. Changing loads and operating profiles can complicate health assessment, while trustworthy sensing and analytics require verification and validation. More collected data does not automatically establish better diagnostic performance.

Most importantly, condition monitoring is not a substitute for machinery risk reduction or hazardous-energy control during servicing. Keep monitoring decisions separate from required safeguards and safe-work procedures.

8. Common Pitfalls and How to Avoid Them

• Buying sensors first: Instrumentation can create data without useful fault coverage. Define the failure mechanism and response decision before selecting hardware.
• Declaring whole machines covered: A monitored bearing does not establish coverage of controls, lubrication, or every mechanical failure. Maintain an explicit coverage register.
• Comparing incompatible operating states: Load or speed changes can be mistaken for deterioration. Preserve operating context and use comparable baselines.
• Confusing two sampling decisions: The interval between inspections differs from the acquisition rate within a measurement. Specify both according to the fault and signal being investigated.
• Leaving alerts without authority: Warnings can remain unresolved when maintenance and production disagree. Assign escalation rights, response expectations, and work-order ownership before launch.
• Overclaiming benefits: Do not label every intervention an avoided breakdown. Separate confirmed findings, estimated avoidance, and observed business outcomes.

9. How ISO 17359 Machine Condition-Monitoring Methodology Relates to Other Frameworks

Before or within the front end: failure analysis and maintenance-policy selection

Failure modes and effects analysis (FMEA) identifies how an item or process can fail and the consequences. Failure modes, effects and criticality analysis (FMECA) adds criticality assessment. Use these to establish what monitoring needs to detect.

Reliability-centered maintenance (RCM) addresses the broader choice of failure-management policies. Prefer RCM when the primary question is which maintenance approach to adopt; use ISO 17359 to develop the monitoring program where condition-based tasks are appropriate.

Alongside: machinery risk assessment

ISO 12100 addresses machinery hazard identification, risk assessment, and risk reduction. Apply that safety discipline alongside monitoring design; a machine-health assessment and a machinery-safety assessment answer different questions.

For deeper technical implementation: diagnostics and prognostics

ISO 13379-1 provides guidance on diagnostic approaches, while ISO 13381-1 addresses prognosis processes. These are useful supplements when decisions require more than detecting an abnormal trend—for example, diagnosing its cause or evaluating future deterioration.

10. Key Takeaways

• Start with machine functions and failure consequences, not available sensors.
• Prioritize asset–failure-mode combinations, especially those affecting bottleneck capacity and quality.
• Require credible warning and enough time to execute a response.
• Connect alerts to ownership, parts, production access, and repair verification.
• Preserve independent safety controls and test assumptions before scaling.

11. FAQs About ISO 17359 Machine Condition-Monitoring Methodology

Does ISO 17359 require continuous monitoring?

No. It considers periodic and continuous monitoring. The appropriate choice depends on fault progression, operating conditions, criticality, and the warning time needed—not a blanket requirement to connect every machine continuously.

Can smaller manufacturers use the methodology?

Yes. Start with a small critical-asset group, a documented monitoring plan, and clearly assigned responses. Portable measurements and specialist analysis can be appropriate alternatives to a plant-wide permanent installation.

Does applying ISO 17359 establish machinery-safety compliance?

No. Its scope is condition-monitoring guidance. Machinery risk reduction and applicable safe-servicing requirements remain separate responsibilities; a healthy-machine reading does not establish that hazards have been adequately controlled.

What should executives ask to see before approving rollout?

Request a failure-mode coverage map, justified measurement methods, representative baselines, response ownership, and pilot findings. Also ask which failures remain undetectable and what alternative controls address them. Approve a functioning decision process, not merely a list of installed sensors.