AI-Based Persona Generation for the Development of a Post-Mission Analyzer Tool for Launch and Re-Entry Operations Artificial intelligence (AI) in the context of safe and efficient air traffic management

Kurzfassung

The increasing number of orbital launches and controlled or uncontrolled re-entries in Europe and worldwide significantly raises the complexity of coordinating space operations with air traffic management (ATM) and other safety-critical domains. Post Mission Analysis (PMA) plays a key role in evaluating the effectiveness, safety, and regulatory compliance of launch and re-entry operations, as well as in improving future mission planning and operational procedures. However, the development of PMA tools faces challenges due to the heterogeneous stakeholder landscape, differing operational perspectives, and varying information needs across national and international organizations. This paper presents an AI-based approach for the systematic creation of operational personas to support the requirements engineering and user-centered design of a Post Mission Analyzer Tool for space launch and re-entry operations. The proposed method leverages large language models (LLMs) and structured domain knowledge to generate realistic, consistent, and traceable personas representing key stakeholders involved in European and international spaceflight operations, including launch operators, air navigation service providers, space surveillance entities, regulatory authorities, and mission coordination centers. The approach combines document-based knowledge extraction, expert-defined constraints, and prompt engineering techniques to ensure that the generated personas reflect real-world roles, responsibilities, decision-making processes, and post-mission information requirements. The resulting personas are used to derive functional and non-functional requirements, identify gaps in existing PMA concepts, and support scenario-based validation of the Post Mission Analyzer Tool. A case study focusing on launch and re-entry operations in the European context demonstrates how AI-generated personas contribute to improved stakeholder coverage, increased consistency in requirements definition, and reduced effort in early systems engineering phases. The results indicate that AI-based persona generation can enhance transparency, reproducibility, and scalability in the development of complex space operations support tools. The presented work contributes to the integration of artificial intelligence into systems engineering methodologies for spaceflight operations and provides a structured foundation for future PMA tools supporting safe and efficient launch and re-entry coordination in increasingly congested airspace and orbital environments.