Due to the rapid expansion of new energy electric vehicles, intelligent and connected electric vehicles, which incorporate artificial intelligence (AI), connection, and energy efficiency, have the benefit of collective intelligence. This makes them well-suited for large-scale urban chores. They have become an essential component in the advancement of social services in smart cities. This study investigates the use of AI-enhanced dispatch algorithms to address job scheduling challenges in urban Electric vehicular Internet of things (EVIOT) -0pln, vbfleets that are intelligent and connected. The primary problems lie in the city’s work allocation methodology and the implementation of AI-driven task execution for each vehicle. AI algorithms that calculate the regional advantages derived from vehicle trajectories are necessary for implementing fleet dispatch systems. This integration guarantees the successful fulfilment of tasks and the return of the vehicle, taking into account the limitations of battery power. The combination of group dispatch techniques and individual AI-based route planning results in a complex task that is NP-hard. This problem involves weighted bipartite graph matching and the travelling salesman problem, both of which are solved using AI. A vehicle dispatch method utilizing artificial intelligence and maximum weight matching has been created to address these obstacles. The system initially utilises a greedy approach to allocate work segments to individual vehicles in sub regions, making use of artificial intelligence for decision-making. Afterwards, it utilises the advantages specific to each region that are produced by analysing vehicle paths using artificial intelligence, in order to create the most effective plan for dispatching cars to sub regions. This approach aims to maximize the overall benefit for the entire region. The efficacy of an artificial intelligence (AI)-driven Programme was evaluated by analysing a 30-day dataset obtained from 238 intelligent sanitation vehicles. The experimental findings demonstrate that the integration of artificial intelligence (AI) into the methodology led to an average improvement of 11.2% in the urban road cleaning rate, as compared to conventional methods.