Engineering Data Funnel

Neuro-symbolic AI for automation technology

Authors

  • Nicolai Schoch ABB
  • Mohamed Elscheikh ABB
  • Mario Hoernicke ABB
  • Sebastian Palacio ABB
  • Nika Strem ABB
  • Katharina Stark ABB

DOI:

https://doi.org/10.17560/atp.v67i10.2805

Keywords:

KI-unterstütztes Engineering, Neurosymbiotische KI, Anlagen-Design-Spezifikation

Abstract

In automation engineering, today, there are many software applications available, but a major challenge is to consistently and coherently transform unstructured multimodal technical design specifications, so that the
existing engineering applications can further process them. Here, firstly, we analyze indepth the requirements for this data transformation, and then introduce the “Engineering
Data Funnel” (EDF), a neuro-symbolic AI, which is based on an agentic AI workflow, in which specialized expert models process multimodal data in mutual coordination and under consideration of domain knowledge. Finally, the EDF is evaluated using examples from various industry segments.

Published

2025-11-03

Issue

Section

Article / Peer Review

Most read articles by the same author(s)

1 2 3 > >>