• Project number: F 2611
• Institution: Federal Institute for Occupational Safety and Health (BAuA)
• Status: Ongoing Project
• Planned end: 2028-03-31

Description:

The digital transformation is driving increasing automation of planning and decision-making processes in the workplace. In particular, within the field of algorithmic management, i.e., digital methods for organising and steering work, there is a growing need for systems that are not only efficient but also transparent, fair, and human-centred. In safety-critical domains such as clinical care, this is especially important, as staffing decisions and workload directly affect the safety and health of employees.

Current AI systems often operate as black boxes: they optimise decisions based on hidden assumptions, limiting trust, acceptance, and participation. They also risk creating structural overload for individual employees. This project therefore investigates how AI-based task allocation systems can be designed to be transparent and explainable. As part of a feasibility study, a knowledge-based AI system will be developed to handle small-scale planning and decision problems and to examine the explainability of its decisions. In parallel, generative AI models will be used to analyse differences in rule compliance, depth of explanation, and trustworthiness.

By comparing knowledge-based and generative approaches, the project makes a significant contribution to algorithmic transparency in AI-supported work systems. It strengthens employee interests, helps prevent work overload, and increases organisational reliability and accountability.