How do you train leadership when seconds count, information is incomplete, and decisions must be made under stress? The collaborative project FEIR – Leadership Training for Emergency Personnel Using Intelligent Virtual Realities develops a VR-based training system for emergency and command personnel in civil protection. Until now, training scenarios were typically assessed subjectively by instructors, without structured, individually traceable evaluation of learning progress. In realistic virtual emergency scenarios, trainees practice tactical decision-making, situational assessment, and safety awareness without facing real-world risk.
What sets FEIR apart: training scenarios are generated automatically and modularly, offered across multiple difficulty levels so that exercises remain reproducible yet varied. At the same time, the system objectively analyzes trainee behavior — for example through paths taken, time required, hazards identified, checkpoints reached, and recall performance. Physiological strain is additionally captured via vital sensors and evaluated using AI-based methods, allowing stress responses to be assessed in relation to demonstrated performance.
The result is a comprehensive picture combining behavioral safety, tactical performance, and physiological strain. Workshops and field evaluations with volunteer fire departments confirmed both the accuracy of the automated assessments and their didactic value for debriefing. FEIR thus combines virtual reality, data analysis, and artificial intelligence into a practice-oriented training tool that prepares emergency personnel for complex, high-stress situations.