Published January 21, 2023 | Version 7
Other Open

Checklist – Measurement and data planning for machine learning in assembly

  • 1. Lab for Measurement Technology, Saarland University
  • 2. Chair of Assembly Systems, Saarland University

Description

Data quality plays a decisive role in fully exploiting the potential of artificial intelligence or machine learning in industry, especially in assembly. The checklist provided here is intended to support industrial users, especially in medium-sized companies, in acquiring high-quality data to minimize the effort required for data analysis and increase the meaningfulness of the results. It supports the planning of a project to use machine learning at an existing assembly plant (brownfield). Therefore, the focus is not on acquiring a new plant or data planning for prototypes in product development. Nevertheless, the checklist can also provide guidance in these use cases. It follows the approach of recording valid data as precisely as possible by contributing expert knowledge and making it available clearly. Due to the high diversity of available machine learning methods, data analysis is only covered at a basic level within the scope of this document.

The primary target groups of this checklist are manufacturing SMEs and large companies.

This checklist is also available in German on the link https://zenodo.org/record/6943476#.Y8-jiXbMJD8.

Notes

Developed as part of a cooperation between the Lab for Measurement Technology and the Chair of Assembly Systems within the framework of the ERDF project "Messtechnisch gestützte Montage" and the follow-up project" "iTecPro – Erforschung und Entwicklung von innovativen Prozessen und Technologien für die Produktion der Zukunft". Future development is carried out in the project "NFDI4Ing – the National Research Data Infrastructure for Engineering Sciences", funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 442146713

Files

20230121_Checklist_V7_Zenodo_EN.pdf

Files (495.5 kB)

Name Size Download all
md5:f2b6d608199eeb814fa228bf246c126c
495.5 kB Preview Download