XAPS: Explainable AI for Automated Production System

Funded by the Federal Ministry of Education and Research

Project Description

Automated Production Systems (aPS) arrange operating resources in complex processes. These resources process intermediate products and check their quality. Depending on the inspection result, an intermediate product is passed on in the process, scrapped or discharged for manual inspection and rework. Although the individual steps are monitored, there are such complex interdependencies between material, processing steps, operating equipment, equipment maintenance and inspection results, that that even if the intermediate products are supposedly processed correctly, the end product may lack the required quality, resulting in expensive rework or even rejects.

The goal of the project XAPS is to develop methods and tools for identifying failures in automated production systems early in the design and operation and guide engineers and technicians to avoid these failures in the future. In this project the consortium combines methods of data science and model-based software engineering to create explanations for possible reasons for failures.


  • Old World Computing GmbH
  • iTAC Software AG
  • Hella GmbH & Co. KGaA
  • Fraunhofer ISST
  • Unviersity of Stuttgart

Project Info

Run Time: 01/2020-12/2022

Funding Agency: Federal Ministry of Education and Research