App-based effort reduction during adaptive testing in the production of variants

 

The aim of the AiF research project "APProVe" is the development of an automated process in an app to reduce the effort of adaptive test planning in variant production. In variant production, a 100% inspection with complete inspection of all inspection characteristics based on small quantities is common. In large-volume production, this inspection effort is prevented by an adaptive random sample inspection, which only applies to key characteristics. The implementation of this random sample inspection often requires companies to have non-existent personnel capacities and statistical knowledge.

The solution to reduce these efforts is to develop algorithms based on statistical methods to define key characteristics and mixed lots. They use historical measurement data of the variants as input. Key characteristics are defined by the first algorithm for each variant identifying inspection characteristics with redundant inspection statements and then defining the production-critical inspection characteristics as key characteristics. The second algorithm uses the key characteristics and compares them across variants based on their measurement data. If they are similar in terms of production technology and statistics, they are grouped into mixed lots. These serve as input for adaptive sampling inspection plans, which adjust the inspection frequency and inspection scope according to the quality level and the risk.

The result is a procedure for inspection planning of an adaptive inspection in variant production. Algorithms for the identification of key characteristics and for the creation of mixed lots as well as sample inspection plans adapted for variant production with risk assessment are integrated in this process. The procedure is implemented as an app that supports the user in the application. The benefits for SMEs result from an app that increases competitiveness through adaptive random sampling, relieved testing personnel, increased resource efficiency and reduced production costs.

Further information can be found on the website of the research project APProVe.