Field testing – The bridge from product development to optimized maintenance

Uptime Engineering > blog > Field testing – The bridge from product development to optimized maintenance
Felddatenanalyse, Field testing

Field tests ask about the unknown risks of future product use. They provide the first real load-response behavior in the product life cycle. This is just as useful for verifying development results as it is for condition-based and predictive maintenance. Especially for durable B2B products and fleets – such as railways, construction machinery, etc. – field tests are an essential building block for reliability and durability.


The promise of virtualization is the elimination of the complex and costly testing programs. However, development practice shows that field tests’ contributions are indispensable for numerous failure risks.  The a priori known failure risks are addressed by focused tests in order to raise the components and subsystems to the highest possible degree of maturity. However, surprises in the field, resulting from novel failure mechanisms or unexpected operating load cases, are not covered. These issues require a complementary “agnostic approach”, which refrains from concrete damage mechanisms aiming at the long-term system behavior under conditions that are as heterogeneous but realistic as possible.

Use field testing to avoid unpleasant surprises 

  • Field tests are useful as a conclusion to system validation if they are carried out under realistic conditions. Then they are representative and ask about all risks – primarily about the hitherto unknown ones.
  • Field tests should therefore not be “accelerated” (HAST, HALT), i.e. they should not address a certain subset of the known risks. They should rather be representative for the future usage cases. Thus, they cover only about 8,000 hours of future operation per year.
  • For B2B applications (railway, construction machinery), the available test duration is much shorter than the lifetime targets. The field test contribution to reliability demonstration is therefore limited. This is acceptable, because it is not the primary objective of field tests.
  • Field tests are designed to detect what all previous activities cannot uncover. Therefore, no failures due to known types of damage must occur in field tests, because these must have already been detected and eliminated in the upstream actions. Field tests therefore also determine the quality of the development process.
  • Therefore, we recommend to postpone field tests until a sufficiently high level of reliability has been proven with simulation and bench tests. Start with mature components and go on with (some) tests beyond the SoP to cover the lifetime targets.

Determine the degradations for lifetime calculation 

The long overhaul/warranty/lifetimes of B2B products cannot be covered by field tests for most failure risks. However, field tests contain a lot of information about the  actual damage under realistic load conditions. Evaluate this information systematically:

  • Measure the load variables that are relevant for the various damage mechanisms (wear rate, thermal aging, fatigue, etc.).
  • After field testing, test critical components on the bench until the end of their life. Evaluate the initial – if feasible the intermediate – and the final states of components.
  • Calibrate the lifetime models with these measured values and evaluate the real load capacity of the components.

For less than 10% of the test costs, these analyses provide you with an – otherwise unavailable – basis for the lifetime calculation and also for the calibration of a risk-focused fleet monitoring. 

Monitor the fleet to prevent unplanned outages 

Degradation generally leads to a gradual change in the system load behavior. Damage indicators include component temperatures, system efficiency, load response, etc. Prevent unplanned outages by observing these indicators.

  • Identify which indicators are useful for which failure risks (based on a failure potential analysis). Monitor these time series already during endurance runs. This creates a robust reference database for the expected behavior of the product.
  • Analyze the history of the indicators (e.g. with the Uptime HARVEST software) to identify degradation.
  • Remove suspicious products from operation before they fail. Check whether the degradation can be explained by the load history or whether a quality problem has been detected by the indicator.
  • Activate this monitoring to manage preventive fleet maintenance (CBM, PDM).


Uptime Pfeil Icon

Evaluation of the efficiency of the development process

Uptime Pfeil Icon

Avoidance of serial damage in the field

Uptime Pfeil Icon

Triggers for condition-based and predictive fleet maintenance, CMB and PDM

Related Posts

Is predictive maintenance a useful concept?

“Operate systems until there is a machine crash?” or “Try to predict the remaining lifespan...

Data quality for maintenance optimization

The quality of current time series data is often inadequate for analytics for maintenance optimization....

The knowledge base for development and maintenance

The knowledge of developers and maintenance technicians is the decisive competitive factor, especially for high-tech...