The paper presents a framework driven by microsimulation to assess the performance of connected automated vehicles at intersection designs. The research explores the behaviour of connected automated vehicles in mixed traffic, focusing on calibrating model parameters to align simulated data with entry capacity functions for various deployment rates of smart vehicles. A suburban signalized intersection in Poland serves as the baseline case study for modelling the transition to a knee-turbo variant in Aimsun. The methodology employs a validation-focused approach that highlights parallels in entry mechanisms at the tested roundabouts. Findings reveal that the performance of connected automated vehicles is significantly affected by intersection geometry and the deployment rates of these vehicles, providing insights for infrastructure project assessments and support the integration of road designs with smart driving technologies.