Project Time4CPS: Time for Cyber Physical Systems

Welcome to the website of the Time4CPS project. Here you will find information about the project as well as current event information and contact persons.

Challenge

The application of artificial intelligence is currently a key innovation driver for production and logistics processes. Many methods exist that can be used to analyze process data. However, projects often fail in their application because an elementary piece of information is not used in production and logistics processes: time. Yet it is precisely the time behavior that reflects many effects and is often available without the use of additional sensors. There is also great potential for process optimization here due to the now standardized time synchronization of all IT and automation systems. So why is this freely available, invisible sensor time so little used in the context of optimization? The main reason is that durations can only be measured between two events.  And in such systems, these events are generally not defined. It is true that measurable events exist, such as control signals to switch on a robot. But usually just the events relevant for optimization are hidden by complex patterns in mostly continuous, interdependent and high-dimensional sensor value sequences and therefore unknown and not explicitly usable for optimization algorithms.

 

Solution approach

The research project therefore aims to develop a methodology and a software platform that automatically discretizes relevant events from typical logistics and production data, which are then used for system monitoring and optimization. The commercial exploitation within this project will be done by the SME Recogizer Analytics GmbH, which is interested in implementing such a methodology. The approach will be evaluated by ISI Automation GmbH & Co. KG, as well as Hendricks Automotive Group and GTP Schäfer Gießtechnische Produkte.  These companies would like to apply the solutions to be developed, for example, to the following optimization scenarios in the area of inter- and intra-logistics and to processes in discrete manufacturing (e.g., at automotive suppliers): 

  1. Varying time durations in logistics processes such as storage or transport often indicate suboptimal automation strategies. 
  2. Dependencies between production durations and product properties allow optimization of production sequences. 
  3. Unexpected time delays in logistics or production are often an indicator of production problems and can even point to quality problems.
  4. Correlations between raw materials or output products and time durations can indicate problems at supplier companies.

Profile

Project title: Time4CPS
Runtime: 24 months
Promotion: Bundesministerium für Bildung und Forschung (BMBF)
Project partner: DISI Automation GmbH & Co. KG
Hendricks Automotive Group
GTP Schäfer Gießtechnische Produkte
Target: Time-intensive process monitoring for cyberphysical systems