Project

myRPA

Experience-based Robotic Process Automation for Knowledge-based Personal Assistants

Robotic Process Automation (RPA) enables further automation of processes across existing systems. As RPA can act like humans and thus is able to read, transfer, and enter data in graphical user interfaces, for example, it allows for automating previously inaccessible areas. RPA is often static and predefined, lacking integration of (procedural) experience and the assistance of employees with appropriate process parts in the respective work context.

Therefore, the goal of the joint project "myRPA" is to reach the next development stage of AI-based assistance systems for the support of knowledge and information workers. Such assistance systems are intended to provide employees with process information on demand through personal semantic support, freeing them from tedious routine tasks by automatically executing workflows across different application systems on the desktop according to need and context. Self-learning mechanisms are used to continuously develop and optimize the workflows. This can improve the efficiency and quality of information and knowledge work and reduce the cognitive load of employees regarding "tedious activities", thus creating more mental freedom for creative work.

This is achieved by combining various AI technologies for knowledge-based assistance (semantic technologies, experience-based systems, and machine learning) with RPA software technology. The combination of both approaches has the potential to realize a (pro-) active, self-learning "information butler" that is able to suggest appropriate actions for the individual work situation in information- and knowledge-intensive activities.

This expands the current range of applications for RPA since, by combining it with knowledge-based assistance, fields of application with significant automatable portions in weakly structured processes can also be supported. From the perspective of knowledge-based assistance, the capture and automatic execution of standard processes by RPA offers a significant workload reduction for users, even for information- and knowledge-intensive fields. The specific details of operating the application systems no longer have to be permanently present to the user in addition to the necessary specialized knowledge, which minimizes tedious work portions and can significantly increase work efficiency. From the point of view of future users in business and public administration, these advantages have a direct effect on increasing efficiency, both for routine activities and for workflows that are performed less frequently. For these, process compliance can also be improved and error rates reduced. Information transfer and process automation go hand in hand, relieving employees and increasing satisfaction. Overall, this results in a significant increase in benefits and thus a significantly increased attractiveness of a myRPA solution on the market.

Partners

  • AmdoSoft Systems GmbH (Koordinator)
  • Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI) (Forschungspartner)
  • envia Mitteldeutsche Energie AG (enviaM) (Anwendungspartner)
  • Licht- und Kraftwerke Helmbrechts GmbH (LuK) (Anwendungspartner)

Sponsors

BMBF - Federal Ministry of Education and Research

01IS21030B

KMU-innovativ

BMBF - Federal Ministry of Education and Research

Publications about the project

Christian Zeyen, Rudolf Koch, Sven Schwarz, Heiko Maus, Ralph Bergmann

In: ICCBR POCBR'22: Workshop on Process-Oriented Case-based Reasoning at ICCBR-2022. Workshop on Process-Oriented Case-Based Reasoning (POCBR-2022) located at 30th International Conference on Case-Based Reasoning (ICCBR 2022) September 12-12 Nancy France CEUR-WS.org 9/2022.

To the publication

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz