Purpose Measurement uncertainty is present in all measurement processes in the field of production engineering. However, this uncertainty should be minimized to avoid erroneous decisions. Present methods to determine the measurement uncertainty are either only applicable to certain processes and do not lead to valid results in general or require a high effort in their application. To optimize the costs and benefits of the measurement uncertainty determination, a method has to be developed which is valid in general and easy to apply. The paper aims to discuss these issues. Design/methodology/approach This paper presents a new technique for determining the measurement uncertainty of complex measurement processes. The approximation capability of artificial neural networks with one hidden layer is proven for continuous functions and represents the basis for a method for determining a measurement model for continuous measurement values. Findings As this method does not require any previous knowledge or expertise, it is easy to apply to any measurement process with a continuous output. Using the model equation for the measurement values obtained by the neural network, the measurement uncertainty can be derived using common methods, like the Guide to the expression of uncertainty in measurement. Moreover, a method for evaluating the model performance is presented. By comparing measured values with the output of the neural network, a range in which the model is valid can be established. Combining the evaluation process with the modelling itself, the model can be improved with no further effort. Originality/value The developed method simplifies the design of neural networks in general and the modelling for the determination of measurement uncertainty in particular.

No abstract

No abstract

PurposeTo ensure product quality within a manufacturing process, inspection processes are indispensable. One task of inspection planning is the selection of inspection characteristics. For optimization of costs and benefits, key characteristics can be defined by which the product quality can be checked with sufficient accuracy. The manual selection of key characteristics requires substantial planning effort and becomes uneconomic if many product variants prevail. This paper, therefore, aims to show a method for the efficient determination of key characteristics.Design/methodology/approachThe authors present a novel Algorithm for the Selection of Key Characteristics (ASKC) based on an auto-encoder and a risk analysis. Given historical measurement data and tolerances, the algorithm clusters characteristics with redundant information and selects key characteristics based on a risk assessment. The authors compare ASKC with the algorithm Principal Feature Analysis (PFA) using artificial and historical measurement data.FindingsThe authors find that ASKC delivers superior results than PFA. Findings show that the algorithms enable the cost-efficient selection of key characteristics while maintaining the informative value of the inspection concerning the quality.Originality/valueThis paper fills an identified gap for simplified inspection planning with the method for the efficient selection of key features via ASKC.

No abstract