Reliability is in focus in connection with product development and production, but what happens once the product has been delivered to the customer?

By Susanne Otto

Both in Denmark and internationally, the most successful companies are those skilled at collecting data from observations and intrinsic field failures, analysing data and looping the information back into the development process.

A system for handling feedback from the market can range from complex ”Post Market Surveillance” systems with analysis of trends, which is required for companies that produce medical devices, to a more informal interaction between service department and R&D. At some of these companies, relevant stakeholders meet for example once a week to discuss important matters that need to be addressed immediately, and dealt with in the long term by ensuring that the problem does not recur in new products.

Make use of valuable feedback from the market

The vision is to make it possible for all companies to exploit the wealth that is to be found in feedback from the market. In the performance contract “Reliable product development based on the Physics of Failure”, the objective is to develop a model for the analysis of field data with the aim to looping it back to the development process and thus reduce service and warranty costs.

The ambition is a simple and effective model, which can be used by all companies that are not faced with e.g. regulatory requirements for specific methods for ”Post Market Surveillance”. The model is based on a study of how five selected Danish and international companies handle and manage feedback from the market.

Both those companies which have only just got their product onto the market and those with mature products are investigated. The companies’ systems for the handling and analysis of products that have failed in the field are studied, inclusive of the ability to use the results in subsequent product development. The potential for improved user information is also studied in connection with field failure by implementing software and/or sensors in products.

New model for failure analysis

The model will consist of two parts; a process for the handling and management of complaints, and how the information in these can be looped back into the development process, as well as a process for systematically finding the root cause or critical failure mechanisms (Physics of Failure).

Practice has shown that classic failure analysis methods such as SEM (Scanning Electron Microscopy), X-Ray, EDS (Electron Dispersion Spectrum) and SAM (Scanning Acoustic Microscopy) are necessary but not always sufficient for determining the critical failure mechanisms in complex problems. Here it is necessary to expand on this with a systematic process for evaluating the results of the analysis and possibly supplement it with expert advice within specialist areas.

This was the case in one example involving a sensor for measuring the salinity of water that failed during use. Countless analyses and tests had been performed, just as there had been a number of modifications made to the sensor, without the manufacturer having been able to achieve an adequate service life when the sensor was subjected to warm water in the use environment. After a systematic process based on methods such as Fishbone and 5 Whys supplemented with further failure analysis, it was discovered that hydrolysis was the critical failure mechanism, and armed with this knowledge, the company was then able to optimise the type of plastic used and thereby avoid the problem.

The processes that are developed in connection with the project are verified in demonstration projects for companies that would like to enter into a collaboration.