Data-driven Evolutionary Optimization of eVTOL Design Concepts Based on Multi-agent Transport Simulations

Electric Vertical Take-Off and Landing (eVTOL) aircraft design concepts are currently developed by many companies and research consortia. An essential topic in the development process is finding the optimal vehicle specification very early on to identify the designs that promise the best fleet operation profit. This paper proposes a novel method that combines open vehicle design concepts with an Evolutionary Algorithm (EA) optimization scheme to find the optimal aircraft specifications, including capacity, range, cruise speed, and height. The proposed optimization framework is evaluated using an activity-based large-scale multi-agent transport simulation for the region of Corsica. The results show that the framework yields different design concepts for different Urban Air Mobility (UAM) network designs. A significant increase in the expected profit can be obtained by adapting the vehicle specification for specific market conditions. This shows the potential of the proposed method, which, through its generic character, could be applied in the future to determine optimal configurations of different products and services for particular markets by maximizing the profit of the fleet operator and the overall utility of the customers.