<p>This study investigates the optimization of urban surveillance systems through the application of Edge Computing Environment and advanced maintenance strategies, aiming to enhance system reliability and efficiency in smart cities. By employing Stochastic Petri Net models, the research evaluates different maintenance approaches including reactive, autonomous, and preventive strategies. Reactive maintenance initiates repairs post-failure, while autonomous systems utilize self-repair mechanisms to minimize downtime. Preventive maintenance incorporates scheduled interventions to reduce system failures caused by aging and wear. This work presents a comprehensive framework for evaluating the dependability of urban surveillance systems. It integrates availability and reliability analyses with maintenance strategies, highlighting the critical role of maintenance in ensuring system performance. A key innovation of this study is the use of sensitivity analysis to identify components that significantly impact system availability, guiding the focus on high-impact areas to optimize maintenance efforts. The research findings are substantiated by case studies that demonstrate the practical applications of the proposed models in real-world scenarios, providing valuable insights for urban planners and policymakers. This work contributes to the field by offering a robust methodology for enhancing the dependability of urban surveillance systems, crucial for maintaining public safety and efficient urban management. The integration of advanced maintenance strategies not only improves system resilience but also offers a cost-effective solution to the challenges of continuous urban surveillance, ensuring sustainable and reliable operation in smart cities.</p>