| Abstract [eng] |
The master’s thesis explores the possibilities of applying artificial intelligence (AI) to short-term car-rental processes in order to automate vehicle damage detection and address both operational efficiency and ethical challenges. The theoretical section identifies the core rental processes—booking, pick-up, return and damage assessment—and shows how AI technologies such as computer vision and machine learning can transform them, delivering greater accuracy, consistency and customer trust. In the empirical section, eight semi-structured interviews were conducted with employees of the Hertz Lithuania branch. Thematic qualitative analysis revealed that AI shortened vehicle inspection time by 25-40 %, reduced manual labour requirements by 10-20 % and halved the number of disputes with customers. Nevertheless, two critical issues emerged: (1) system accuracy under poor lighting conditions and (2) customers’ mistrust of “black-box” decisions. The thesis therefore recommends: implementing Explainable AI visualisations that clearly show how the algorithm identifies damage; adapting algorithms to various lighting conditions and using 3-D imaging methods for complex defects; integrating IoT sensors for continuous vehicle-condition monitoring; upgrading staff competences so they can operate and supervise AI systems, and establishing a transparent data-management policy. The scientific novelty lies in analysing the synergy between AI and process management in the car-rental sector, a topic that has so far received little attention. The practical value is a set of guidelines showing how short-term rental companies can integrate AI solutions effectively and ethically, bridging the gap between theoretical models and real-world business practice. The study contributes to the broader discussion on sustainable AI development in mobility services, demonstrating that transparent, customer-understandable algorithm deployment is essential for the sector’s competitiveness and long-term growth. |
