Vehicular transportation is one of the most widely used modes in modern cities for reaching key destinations such as workplaces, healthcare facilities, recreational areas, and educational institutions, among others. However, the performance of vehicular traffic on these roads can vary significantly due to the influence of different environmental variables. In the literature, factors such as traffic incidents, weather conditions, road infrastructure, and driving habits, among others, have been identified as impacting vehicular traffic performance. In this context, predictive models have been developed to anticipate congestion at specific points in cities based on statistics, machine learning, simulation, and complex networks. This study proposes a novel index aimed at assessing the level of performance of vehicular traffic on streets based on the relationship among relevant urban environmental variables. This index is generated through Genetic Programming, considering a set of variables related to traffic, incidents, and services. The case study will focus on the streets of the Tlalpan Municipality in Mexico City.