Reliability metrics

Sep 17, 2020 / by GM International

Topics: Safety news

Reliability plays a key role in the cost-effectiveness of systems and is defined in terms of probability or frequency through a set of statistical indicators that cannot be ignored.

In engineering and safety systems, reliability determines the estimation, prevention and management of durability, uncertainty and failure levels of a machine or a plant. Reliability metrics are statistical in nature and are not a single and comprehensive tool for determining the cause of a failure, and they can’t be used to predict reliability with certainty and accuracy.

On the other hand, reliability engineering and safety techniques are closely related because they use common analysis methods. Reliability engineering focuses on the costs of failure due to downtime of the system, on the cost of spare parts, equipment under repair, personnel and warranty rights.

Security technologies normally focus on protecting people. They only deal with particularly dangerous failure factors. High reliability is a safety factor, the results of which also depend on the design and production methods and not only on statistical-mathematical metrics and models.


Reliability KPIs

Industrial and plant reliability is based on KPIs that help monitoring activities of maintenance over time. These indicators show whether the company or department operate within acceptable levels. Not only they provide insight on the performance of the maintenance department, but they are also displayed via dashboards or other computer media, allowing the company to take decisions about asset reliability, budget and production planning. Let’s see the main ones.



The MTBF (Mean Time Between Failures) indicator expresses a statistical measurement of the expected average time between failures. It represents the average time between the occurrence of one fault and the beginning of the next one. Very common in maintenance and asset management, MTBF is usually measured in hours. It is a statistical indicator that helps plant or maintenance managers understanding and preventing the operation of a machine or an entire plant.

Since MTBF indicates the operating time of an asset, it is a metric that is closely related to the calculation of the reliability of plant and equipment, as much as metrics used to calculate the operating availability of the machine, in particular OEE (Overall Equipment Effectiveness).

These features make MTBF one of the main KPIs, the monitoring of which is one of the key factors to increase the productivity of a plant and to measure the reliability of an asset.



Mean Time To Repair or MTTR is the average time an asset (instrument, machinery, plant) needs to become operational again. In its simplest formulation it is the ratio between total maintenance time and number of repairs. It is a fundamental measure of the maintainability of repairable elements. Mathematically expressed with a value representing an average time, low values indicate better performances. The aim of the maintenance department is to keep the MTTR within acceptable values, to reduce them as much as possible. Tracking and monitoring MTTR can help reducing repair time, improving warehouse management and taking informed decisions about maintenance strategies.



Within the same parameters, we also find MTTF (Mean Time To Failure) which refers to the average time of the occurrence of a failure. MTBF and MTTF are both statistical parameters used to represent the reliability of components that shouldn’t be confused with the average service duration. MTTF values are typically taken into consideration for instrumentation and sensors, usually replaced in case of failure. It is important to note that MTBF is the sum of MTTF and MTTR. Therefore, if a failure occurs at time t1, it will take time for the repair to be completed and time for the failure to occur after time t2.

In substance, MTBF = t2 - t1 = MTTF + MTTR. Attention should be used in order to avoid confusing MTBF value with failure rate, which instead represents the number of assets not working, after a certain time among the total number of those monitored. To this analysis it is necessary to add the formula for Reliability = MTBF / (MTBF + MTTR). This equation establishes a relationship between the improvement of MTBF and other metrics related to plant performance.



AVA (Availability) and PMC (Planned Maintenance Percentage) are not to be overlooked. Both are expressed as percentages. AVA is an indicator of the productivity of an asset and the main component of the OEE that measures the total efficiency of an asset and summarizes its availability, efficiency and quality. PMC is the percentage of maintenance hours that are necessary for planned maintenance activities compared to unplanned maintenance. It is calculated dividing planned maintenance work hours over the total maintenance hours of a precise period of time.


New call-to-action