This study investigates the relationship between road geometry and driver workload, with a focus on Driver Eye Blink Rate (EBR) as a key indicator of cognitive load. Using an AI driven approach, advanced computer vision techniques are employed to automate blink detection from driver camera footage. Facial landmark models are applied to calculate the Eye Aspect Ratio (EAR), which serves as the foundation for identifying blink events. Road segments, such as curves, tangents, and transitions, are analysed through colour coded durations extracted from annotated Excel sheets, allowing for a detailed exploration of how these features influence driver attention. By correlating EBR data with road geometry characteristics, the study uncovers patterns and associations that provide valuable insights into the relationship between road design and cognitive workload. The methodology offers a robust framework for real time monitoring and assessment, with results organized into comprehensive data outputs for further analysis. These findings highlight the potential of AI enhanced systems to improve the precision and scalability of workload assessment, ultimately contributing to safer and more efficient road design strategies.