November 2019

Metrologie für heterogene Sensornetzwerke und Industrie 4.0
(tm - Technisches Messen, Band 86, Heft 11)

Networks of sensors for different measured variables increasingly represent the backbone for a multitude of application areas, for example in industry, mechanical engineering and environmental monitoring. The merging of data (sensor fusion) plays a central role in the application and is generally a well investigated field of research. However, the consideration of metrological basic principles such as calibration, measurement uncertainties and thus traceability to the SI unit system for comparable and reproducible measurement results is comparatively little investigated. 

August 2018


The ZeMA testbed was developed for condition monitoring, lifetime prognoses and end-of-line tests of electromechanical cylinders (EMCs) with a spindle drive. Long-term high load and speed driving tests are carried out at the testbed until an EMC failure occurs. Based on this, relevant sensors and characteristic signal patterns can be identified for condition monitoring as well as residual lifetime estimation of the EMCs.

Conference Talks

  • "Mathematical framework for metrology in the factory of the future", MathMet 2019
  • "Redundant information in sensor networks and uncertainty quantification", MathMet 2019
  • "Influence of synchronization within a sensor system on machine learning results", MathMet 2019
  • "Influence of synchronization within a sensor system on machine learning results", Congres International de Metrologie 2019
  • "Assuring measurement traceability to ATE systems for MEMS temperature sensors testing", Congres International de Metrologie 2019
  • "Methods for dynamic calibration and augmentation of digital acceleration MEMS sensors", Congres International de Metrologie 2019
  • "Calibration 4.0 – Information system for usage of digital calibration certificates", Congres International de Metrologie 2019
  • "Metrology for the Factory of the Future: towards a Case Study in Condition Monitoring", IEEE I2MTC 2019
  • "Decentralized software development facilitating CI/CD to produce high quality, open-source code in a European metrological joint research project", deRSE19 - Conference for Research Software Engineers in Germany 2019
  • "Multi Agent System for Machine Learning Under Uncertainty in Cyber Physical Manufacturing System", SOHOMA 2019
  • "A smart sensor concept for traceable dynamic measurements", XXII IMEKO World Congress 2018

Data sets

In order to support the development of machine learning methods in metrology, the project publishes selected data sets at

Already published data sets:



An important part of the project is the development of software implementing the developed methods. This software is open-source and hosted at



During the course of the project several videos have been recorded for webinars, tutorials and virtual conferences. These videos can watched here.

Currently, the following videos are available.

Presentations at IEEE MetroInd4.0 2020

  • "From dynamic measurement uncertainties to the Internet of Things and Industry 4.0" (S. Eichstädt, PTB)
  • "Quantifying metrological redundancy in an Industry 4.0 Environment" (G. Kok, VSL)
  • "Uncertainty evaluation for metrologically redundant sensor networks" (G. Kok, VSL)

Presentations at Met4FoF Stakeholder Workshop

  • "Dynamic calibration of digital output sensors - concepts for primary methods" (Th. Bruns, PTB)
  • "Dynamic calibration of digital output sensors - implementation" (B. Seeger, PTB)
  • "Provisioning of measurement traceability for MEMS temperature sensors" (V. Fernicola, INRIM and E. Tamburini, SPEA)
  • "Simulation for assessment of ATE testing" (J. Sousa, IPQ)
  • "Advanced forming research centre - radial forging testbed" (G. Gourlay, AFRC)
  • "Redundant measurements in the factory of the future" (G. Kok, VSL)
  • "Uncertainty quantification with machine learning and multi-agent system" (Y. B. Xiang, CAM)
  • "Uncertainty propagation for feature extraction" (T. Dorst, ZEMA)
  • "Bayesian feature-level sensor fusion" (L. Coquelin, LNE)
  • "Bayesian approach to account for timing effects" (K. Jagan, L. Wright, P. Harris, NPL)


  • Tutorial on application of machine learning methods with the ZeMA data sets