Development of AI-based algorithms for automating the analysis of work processes in the skilled trades using video data
Project duration: 01/2021 – 06/2022
Process optimisation is a key tool for designing workflows that are efficient in various respects. In industry in particular, increasing automation and digitalisation – along with the accompanying intelligent and digitally networked systems – mean that, through process optimisation methods, individual production steps can increasingly be organised either partially autonomously or completely autonomously. As a result, large sections of industrial value chains have already been optimised in recent years. The skilled trades sector has not yet seen such a development, as the digitalisation of production and manufacturing steps – which are significantly more heterogeneous – is far more difficult to achieve, and this digitalisation is an essential basis for process optimisation. The collaborative project ‘WIR! InnoHandwerk – Analysis and Optimisation of Craft Manufacturing Processes in the Bakery Trade’ addresses this problem by designing, implementing and testing a service concept for economically promising process optimisation in the skilled trades, based on technology available on the market.
The sub-project “Development of AI-based algorithms for the automation of work process analyses in the craft sector using video data” at FTZ Leipzig focuses on (a) the implementation of a wireless multi-camera system for recording work processes in the craft sector and (b) the automated analysis of the video data recorded in this way. In particular, AI-based methods for the detection, identification and tracking of individuals involved in a craft production process are being implemented and evaluated. The data obtained in this way forms the basis for certain process parameters, such as the distances travelled between workstations. In this way, it is possible, amongst other things, to estimate the proportion of value-adding activity relative to the total production time. The results of the research project thus contribute to the analysis of work processes within the heterogeneous production environment of small and medium-sized craft businesses.
Contact
Prof. Dr.-Ing. Mirco Fuchs
Chair of Computer Vision and Machine Learning
Telephone: +49 (0)341 3076 3104
Email: mirco(dot)fuchs(at)htwk-leipzig.de
