Discrete manufacturing
Reliability is judged by throughput. A stoppage costs lost production, so maintenance performance is reported as OEE and as MTBF / MTTR per asset, and improvement work is prioritised with Pareto analysis of downtime causes.
| Metric | What it tells you |
|---|---|
| OEE | Availability × performance × quality — where the capacity is lost |
| MTBF / MTTR | How often an asset stops, and how long it stays down |
| Planned maintenance % | Share of maintenance hours that were scheduled rather than reactive |
| PM compliance | Preventive tasks completed on time |
Standards and frameworks: ISO 22400-2 (manufacturing KPIs, including OEE), ISO 55000 series (asset management), Total Productive Maintenance (TPM), and SAE JA1011 for reliability-centred maintenance.
Automotive
Automotive reliability is driven by the supply chain: customers require suppliers to prove process capability and measurement capability before parts ship, and to manage risk through FMEA.
| Requirement | Commonly used threshold |
|---|---|
| Process capability (ongoing) | Cpk ≥ 1.33 |
| Process performance (initial, PPAP) | Ppk ≥ 1.67 |
| Gage R&R (% of total variation) | < 10% acceptable · 10–30% conditional · > 30% not acceptable |
The AIAG & VDA FMEA Handbook (2019) replaced the RPN with Action Priority (AP), which weights severity first — so a high-severity failure mode can need action even with a modest RPN. Customer specifics override these defaults; always check the customer-specific requirements.
Standards and frameworks: IATF 16949, the AIAG core tools (APQP, PPAP, FMEA, MSA, SPC), and the AIAG & VDA FMEA Handbook.
Aerospace & aviation
Safety sets the requirements. Designs are analysed for failure probability per flight hour, and maintenance programmes are built from failure-mode analysis rather than fixed calendars.
| Metric | Use |
|---|---|
| Failure rate per flight hour | Safety assessment against probability objectives |
| MTBUR | Mean time between unscheduled removals — drives spares and maintenance cost |
| Dispatch reliability | Share of departures without a technical delay or cancellation |
| Weibull β | Whether a component is wearing out (β > 1) or failing randomly (β ≈ 1) |
Standards and frameworks: AS9100 (quality management), SAE ARP4754 and ARP4761 (system development and safety assessment), MSG-3 (maintenance programme development), and MIL-HDBK-217F (electronic reliability prediction, still widely referenced).
ToolsWeibull AnalysisReliability R(t)Failure RateSpare Parts
Energy & utilities
Distribution utilities report reliability from the customer's side — how often supply was interrupted and for how long — using the indices defined in IEEE 1366.
Total customer interruptions ÷ total customers servedTotal customer interruption minutes ÷ total customers servedSAIDI ÷ SAIFICAIDI is effectively the network's MTTR as a customer experiences it; SAIFI is the inverse of the time between interruptions. Generation assets are tracked with forced-outage rates, and asset-heavy utilities often organise the whole programme around ISO 55000.
Standards and frameworks: IEEE 1366, NERC reliability standards (North America), ISO 55000, and ISO 14224 for structured equipment failure data.
Data centres & IT
Availability is sold as a promise, so it is expressed in “nines” and in allowed downtime. Physical infrastructure relies on redundancy (N+1, 2N); software services manage reliability with service-level objectives and error budgets.
| Availability | Downtime per year | Downtime per 30 days |
|---|---|---|
| 99% | 87.6 h | 7.2 h |
| 99.9% | 8.76 h | 43.2 min |
| 99.99% | 52.6 min | 4.32 min |
| 99.999% | 5.26 min | 25.9 s |
An SLO of 99.9% over 30 days leaves an error budget of 43.2 minutes; once it is spent, the usual practice is to prioritise reliability work over new releases.
Standards and frameworks: Uptime Institute Tier Standard (Tier I–IV), ANSI/TIA-942 (data centre infrastructure), and site reliability engineering practice for SLOs.
Healthcare equipment
For clinical engineering teams the question is whether a device is available and safe when a patient needs it. Programmes are risk-based: life-support and high-acuity equipment gets tighter inspection and faster response than low-risk devices.
| Metric | Use |
|---|---|
| Equipment uptime | Availability of critical devices during scheduled clinical hours |
| Time to restore (MTTR) | Vendor and in-house service performance |
| Scheduled maintenance completion | Inspections and PM done on time, by risk class |
Standards and frameworks: IEC 60601 (safety of medical electrical equipment), IEC 62353 (recurrent testing after repair), and ISO 14971 (risk management for medical devices). Accreditation bodies set the local requirements for maintenance programmes.