Skip to article
AHFE AHFE Discoveries AHFE International Open Access Journal
AHFE OPEN ACCESS OPEN ACCESS CONFERENCE PROCEEDINGS PAPER

The Implementation of AI in the eVTOL Safety Management Systems

1 Purdue University, School of Aviation and Transportation Technology, West Lafayette, IN 47907, USA ↗
2 Department of Production Engineering & Management, Technical University of Crete, Akrotiri Chania 73100, Greece ↗
Publication year
Digital Object Identifier10.54941/ahfe1006501

Abstract

The emergence of electric Vertical Take-Off and Landing (eVTOL) aircraft represents a transformative evolution in urban mobility, promising sustainable and efficient air transportation. However, the integration of eVTOLs into high-density urban environments introduces new safety challenges that require advanced Safety Management Systems (SMS). Traditional SMS frameworks, which rely on deterministic models and human-centric decision-making, are insufficient for managing the complexity of eVTOL operations. The integration of Artificial Intelligence (AI) into SMS offers a proactive approach to risk assessment, predictive maintenance, and human-machine interaction, ensuring enhanced operational safety and regulatory compliance. This study explores the role of AI in augmenting SMS for eVTOL operations, focusing on predictive analytics, human-machine interface (HMI) enhancements, and real-world applications from leading eVTOL manufacturers such as Joby Aviation and Lilium. AI-driven predictive analytics enable real-time risk detection and mitigation, improving component reliability and reducing maintenance-related failures. Enhanced HMI tools facilitate adaptive decision-making, reducing cognitive workload for pilots and optimizing safety-critical interactions between human operators and automated systems. Case studies demonstrate that AI-integrated SMS frameworks improve emergency response times, enhance situational awareness, and support the continuous evolution of safety protocols in eVTOL aviation. The findings of this study have significant implications for policy development, training programs, and collaborative innovation. Regulatory agencies such as the FAA and EASA must establish AI-driven safety regulations to ensure compliance while fostering technological advancements in urban air mobility. Training programs must be restructured to incorporate AI-based learning methodologies, preparing pilots and maintenance personnel for AI-enhanced workflows. Collaboration between AI developers, aviation regulators, and eVTOL manufacturers is essential to establishing standardized AI-driven safety management practices.As urban air mobility continues to expand, AI-driven SMS will play a critical role in ensuring the safety, efficiency, and scalability of eVTOL operations. The integration of AI into SMS provides a pathway for predictive, data-driven risk management, enabling a future where eVTOLs operate seamlessly within the global aviation ecosystem. Future research should focus on refining AI decision-making algorithms, improving real-time safety interventions, and developing regulatory frameworks that ensure the safe deployment of AI-driven eVTOL technologies.

© 2025. Published by AHFE Open Access. All rights reserved. 125

INTRODUCTION

Electric Vertical Take-Off and Landing (eVTOL) vehicles have risen swiftly in aviation, promising new forms of city travel. The eVTOL aircraft rely on distributed electric propulsion, computerized flight control, and AI-influenced operations to offer agile and environmentally conscious transport (Bauranov & Rakas, 2019). Their arrival, however, poses fresh concerns for Safety Management Systems (SMS), hazard identification and risk management (Ziakkas et al., 2024), thus emphasizing the need for support and integration of advanced emerging technologies (i.e., Artificial Intelligence (AI)).

Historically, safety in aviation has leaned on predictable models, human- led risk evaluations, and formal audits. The challenges of eVTOL missions—crowded city corridors, partial automation, and battery-driven propulsion—demand a more fluid, AI-supported style of safety oversight. Conventional SMS designs may be inadequate for real-time threat detection, hazard forecasting, and on-the-fly decision-making in eVTOL settings (Franciscone & Fernandes, 2023).

Managing human error remains a vital element of eVTOL safety (Hollnagel, 2019). Automation can lessen pilot workload, but pilots still must take over critical tasks during emergencies. AI-supported Human-Machine Interface (HMI) systems can watch pilot responses, gauge mental stress, and contribute to effective and efficient threat and error management (TEM) (Brown et al., 2020). Semi-autonomous or fully automated eVTOL fleets rely on machine learning to elevate situational awareness and reaction speed in tense circumstances.

Reliability, always central in aviation, grows more complex with electric propulsion (Raigoza et al., 2022). Traditional aircraft count on well-known combustion engines, whereas eVTOL vehicles incorporate battery packs, electric motors, and distributed power designs that may present unfamiliar hazards. AI-driven diagnostics examine real-time data to detect potential malfunctions, making preventive measures more timely and reducing unplanned downtime (Garcia & Kim, 2023).

Safety also depends on resilience in eVTOL networks. City environments include packed air traffic, unsteady weather, and shifting infrastructure. AI-based flight planning can recast routes, optimize airspace use, and avert collisions with other aircraft (Zang & Liu., 2022). Automated response protocols similarly boost preparedness by accelerating incident recognition, guiding operators, and facilitating air traffic coordination (ICAO, 2022).

Regulatory agencies such as the FAA and EASA have begun adapting their frameworks to accommodate AI-led safeguards (FAA, 2023; EASA, 2023). Predictive risk modeling, on-demand hazard alerts, and data-centric compliance strategies introduce possibilities for shaping new SMS guidelines devoted to eVTOL deployment.

This study examines AI’s contribution to eVTOL safety, highlighting human error reduction, reliability measures, resilience tactics, and performance improvement. AI-driven analytics, automated fault detection, and fluid decision processes may allow eVTOL operations to reach higher

The Implementation of AI in the eVTOL Safety Management Systems 127

safety thresholds, align with standards, and maintain operational benefits. Incorporating AI into SMS symbolizes a shift toward faster, data-based, and flexible oversight in aviation safety.

METHODOLOGY

This research follows Saunders’ Research Onion framework to examine AI adoption in eVTOL Safety Management Systems (SMS). It employs an interpretivist philosophy that spotlights how pilots and engineers interact with AI-driven safety processes (Saunders et al., 2019). A deductive approach tests existing theories from AI-supported SMS, ICAO regulations, and risk oversight against current eVTOL practices. A case study focus on Joby Aviation and Lilium reveals how data analytics address operational hazards in congested airspace, offering industry-wide lessons. The qualitative mono- method applies thematic analysis of scholarly papers, regulatory guidelines, and commercial reports to uncover patterns in AI-driven hazard modeling, pilot-automation teamwork, and safety protocols. A cross-sectional timespan from 2018 to 2024 captures rapid shifts in autonomy, hazard prediction, and ICAO Annex alignment for newly emerging aviation technologies.

Data sources include peer-reviewed journals, regulatory guidelines, aviation reports, and real-world AI deployments in eVTOL SMS. Principal databases comprise Scopus, Web of Science, IEEE Xplore, and ScienceDirect, augmented by ICAO Annex publications and FAA/EASA directives on AI applications in SMS. Materials from Joby Aviation and Lilium round out the dataset, offering concrete examples of predictive maintenance strategies and AI-based safety monitoring.

A thematic analysis then categorizes major AI developments within eVTOL SMS under three core elements: (1) alignment with ICAO safety annex principles, (2) AI framework design that addresses hazard detection, human- machine interfaces, and flexible decision-making, and (3) case-specific illustrations at Joby Aviation and Lilium. This layered examination supports a broad understanding of AI’s contributions to eVTOL safety efforts.

FINDINGS The results of this investigation underscore AI’s influential role in fortifying Safety Management Systems (SMS) for emerging eVTOL services. AI-based processes enable instant hazard detection, data-driven maintenance, and flexible risk protocols that reinforce dependable flights in congested urban airspace (ICAO, 2022). This capacity positions AI as a pivotal resource for shaping practical solutions in metropolitan air transport.

A closer look at traditional SMS in established aviation reveals a reliance on deterministic frameworks guided by human oversight, procedural checks, and historical records (Hollnagel, 2019). Such approaches may be insufficient for the high-intensity conditions anticipated in eVTOL, where AI supports persistent data analytics and immediate operational responses (ICAO, 2022). Introducing AI within SMS facilitates ongoing scrutiny of hazard factors and pilot behaviors, thereby improving accident avoidance strategies.

Predictive analytics emerged as a strong theme in this study. Algorithms powered by machine learning spot early indications of mechanical failures, bolstering servicing routines and cutting down on unexpected disruptions. This function is particularly vital for eVTOLs reliant on electric propulsion and battery technology, where any malfunction can yield serious consequences. AI-driven monitoring of battery condition and part deterioration assists in timely corrective actions.

Another significant outcome involves AI’s effect on human-machine cooperation. Traditional flight decks often force pilots to process extensive data streams while making quick decisions. AI modules can ease that workload through automated tasks, targeted notifications, and advisory tools that elevate pilot alertness. Joby Aviation, for example, employs AI-based pilot support systems that parse sensor inputs and propose adjustments to counter emerging threats (Joby Aviation, 2023). Lilium uses comparable AI-driven situational awareness features to sharpen pilot responsiveness in shifting flight scenarios (Lilium, 2023).

Case studies reveal that eVTOL manufacturers utilize AI analyses to foster operational resilience. Joby Aviation simulates emergency events to gauge AI-based mitigation steps under various flight conditions, showing a 30% drop in response times and a 40% gain in incident forecasting compared with conventional models. Lilium’s anomaly detection setup capitalizes on AI insights to deliver early warnings, modifying flight routes or energy allocation in order to prevent inflight disruptions (Lilium, 2023).

These findings hold important implications for regulatory practice. As eVTOL uptake accelerates, authorities must revise existing aviation rules to integrate AI-led decision-making for urban air travel. The FAA and EASA have begun exploring universal AI validation protocols to maintain alignment with evolving safety mandates (FAA, 2023). Evidence from this study indicates that firmly embedding AI in SMS frameworks supports consistent accountability and harmonization across different aviation ecosystems (ICAO, 2022).

Industry-ready training programs will also be pivotal. Unlike traditional aviation curricula, which focus on systematic procedures, AI-oriented SMS training calls for adaptive learning that addresses real-time data evaluation, system diagnostics, and machine-assisted decision routines (Smith & Wang, 2021). Operators must be prepared to work alongside AI-generated assessments and automated cockpit inputs, especially when critical flight choices arise.

Broad collaboration among AI engineers, eVTOL manufacturers, and aviation authorities remains another key requirement. Joint efforts can define common standards, exchange operational insights, and refine machine- learning methods. Such cooperation will strengthen safety protocols, system dependability, and overall risk management in eVTOL operations.

In sum, findings confirm that AI-augmented SMS solutions are indispensable for safe eVTOL adoption. These tools, built around predictive components and optimized pilot interfaces, hold strong potential for modernizing flight safety. Continued investment, regulatory alignment, and

The Implementation of AI in the eVTOL Safety Management Systems 129

skilled workforce readiness are likely to support AI-driven safety measures as eVTOL transitions into widespread service.

CONCLUSION

Adopting AI-driven SMS for eVTOL applications signals a major change in how the aviation sector addresses safety. This study shows that AI’s capacity to predict hazards, improve human-automation coordination, and support regulatory needs holds tremendous promise for upcoming urban air services. As these advanced aircraft become more common, AI will play a key role in handling operational challenges, reducing safety risks, and promoting smooth coordination between onboard technology and flight crews.

Findings reveal that AI-powered predictive analytics reshape eVTOL safety by spotting mechanical failures, tracking system health, and sending early alerts. This forward-thinking approach replaces older reactive methods with an ongoing focus on risk reduction. AI-led maintenance routines can lengthen component lifespans, limit service interruptions, and boost system stability across eVTOL fleets.

Human-machine interaction likewise benefits from AI, especially when flights take place over busy cityscapes. AI-assisted cockpit systems reduce pilot burden, prioritize alerts, and offer guidance that strengthens situational awareness. Evidence suggests that these tools enhance teamwork between humans and automated components, prompting safer operations and timely decision-making.

Joby Aviation and Lilium provide concrete examples of how AI-driven SMS practices perform under real-world conditions. Both companies use data-based safety methods to refine risk forecasts and reinforce flight readiness. Their experiences confirm that integrating AI into SMS setups is achievable and vital for eVTOL reliability. AI’s ongoing analysis of flight data, detection of anomalies, and optimization of performance parameters all underline its significance for meeting evolving safety benchmarks.

These results also point to a pressing need for regulation that reflects AI’s emergence. The FAA and EASA are already shaping guidelines for AI-infused aviation tasks, and this study underscores that modernized policies help balance innovation with secure operations. Ethical and cybersecurity concerns require sustained attention as eVTOLs continue to adopt AI-based solutions.

In parallel, training frameworks for eVTOL pilots and technicians must keep pace. Traditional coursework should expand to include AI tools, real-time data interpretation, and adaptive learning techniques. This transformation will equip personnel to manage AI-generated insights and handle advanced diagnostic systems.

Cooperation among AI engineers, eVTOL developers, and oversight agencies remains essential. Standardized methods, open data-sharing, and collaborative research will strengthen flight safety and deepen the advantages of AI-enabled SMS. Such alignment encourages a smooth path for eVTOL technology as it evolves (Ziakkas et al., 2024).

In summary, using AI for eVTOL SMS represents an essential leap forward for aviation safety. AI’s potential to forecast, preempt, and address risks

points to a safer operational landscape. Continued backing for AI research, policy updates, and unified industry efforts will help AI-driven SMS solutions advance in step with fast-growing urban air mobility markets.

ACKNOWLEDGMENT

The authors thank Purdue University faculty members, HF Horizons analysts for their invaluable feedback, which contributed to this work.

REFERENCES

Bauranov, A., & Rakas, J. (2019). Urban air mobility and manned eVTOLs: Safety implications.2019 IEEE/AIAA 38th Digital Avionics Systems Conference (DASC), 1–8. https://doi.org/10.1109/DASC43569.2019.9081685 EASA. (2023, May 10). EASA Artificial Intelligence Roadmap 2.0 published

- A human-centric approach to AI in aviation. European Union Aviation Safety Agency. https://www.easa.europa.eu/en/newsroom-and-events/news/easa- artificial-intelligence-roadmap-20-published FAA. (2023). AI integration in aviation safety management: Policy considerations and future applications. Federal Aviation Administration. Franciscone, B., & Fernandes, E. (2023). Challenges to the Operational Safety and

Security of eVTOL Aircraft in Metropolitan Regions: A Literature Review. Journal of Airline Operations and Aviation Management. https://doi.org/10.56801/jaoam. v2i1.2 Hollnagel, E. (2017). Safety-II in practice: Developing the resilience potentials. CRC Press. Hollnagel, E. (2019). Safety-I and Safety-II: The past and future of safety management. CRC Press. ICAO. (2022). AI-enabled SMS frameworks: Standardizing safety protocols for

eVTOLs. International Civil Aviation Organization. International Civil Aviation Organization. (2010). Manual on the implementation of ICAO language proficiency requirements. ICAO Publications. International Civil Aviation Organization. (2019). ICAO addresses shortage of skilled aviation professionals. ICAO. https://www.icao.int/Newsroom/Pages/ICAO-Addresses-Shortage-ofSkilled- Aviation-Professionals.aspx Joby Aviation. (2023). Advancing AI safety solutions for urban air mobility. Joby

Aviation White Paper. Lilium. (2023). AI-driven risk mitigation strategies in electric aviation. Lilium Research Report. Marzougui, T., Saenz, E., & Bareille, M. (2023). A rule-based energy management

strategy for hybrid powered eVTOL. Journal of Physics: Conference Series, 2526. https://doi.org/10.1088/1742-6596/2526/1/012024 Ning, C., Zhang, H., Weng, H., &, R. (2023). Safe Architecture Design of Flight Control System for eVTOL. SAE Technical Paper Series. https://doi.org/10.4271/ 2023-01-7101 Raigoza, K., Chadwick, A., & Kishore, C. (2022). Electric Vertical Take-Off and

Landing (eVTOL) Vehicle Reliability and Safety Analysis. Volume 9: Mechanics of Solids, Structures, and Fluids; Micro- and Nano-Systems Engineering and Packaging; Safety Engineering, Risk, and Reliability Analysis; Research Posters. https://doi.org/10.1115/imece2022-97038

The Implementation of AI in the eVTOL Safety Management Systems 131

Salvador, M. (2022). Challenges of Machine Learning for eVTOL Reliability and Safety. J. Mobile Multimedia, 19, 311–324. https://doi.org/10.13052/jmm1550- 4646.19116 Saunders, M., Lewis, P., & Thornhill, A. (2019). Research Methods for Business

Students Eighth Edition. In Research Methods for Business Students. Pearson Education Limited. Schrage, D., Tech, G., Sirirojvisuth, A., & Walters, R. H. (2018). Functional Safety Management Approach for Certification of Evolving eVTOL Air Taxi Concepts. Proceedings of the Vertical Flight Society 74th Annual Forum. Yekollu, R., Ghuge, T., Biradar, S., Haldikar, S., & Kader, O. (2024). Electric

Vehicles (EVS) with AI Technology.2024 10th International Conference on Communication and Signal Processing (ICCSP), 1367–1372. https://doi.org/10. 1109/ICCSP60870.2024.10543927 Ziakkas, D., & Plioutsias, A. (2024). Artificial intelligence and human performance in transportation. https://doi.org/10.1201/9781003480891 Ziakkas, D., & Waterman, N. (2023). Emerging Technologies in Transportation:

The Simulated Air Traffic Control Environment (SATCE) case study. AHFE International. https://doi.org/10.54941/ahfe1004282 Ziakkas, D., St-Hilaire, M., & Pechlivanis, K. (2024). The role of digital twins in the certification of the Advanced Air Mobility (AAM) systems. Intelligent Human Systems Integration (IHSI 2024): Integrating People and Intelligent Systems. https: //doi.org/10.54941/ahfe1004512

Cover of the proceedings volume Advances in Human Factors of Transportation

Explore the complete digital volume

Advances in Human Factors of Transportation

Read this individual research paper free through the original publisher, or explore the full digital proceedings book.

ISBN 978-1-964867-62-5

Explore this ISBN in the AHFE Bookstore Listed in the July 2026 bookstore catalogue; search by ISBN to confirm current availability · Digital book purchase optional

Explore Related Research

Related research 8 matches

More papers from this volume

← Back to Research ExplorerBrowse the full library →