Detecting anomalous behavior in real-time for an urban area from large data is a challenging problem. Efficient parking management in urban areas is crucial for optimizing space utilization, improving traffic flow, and enhancing the overall urban experience. This paper presents a comprehensive study on the application of time series analysis techniques for detecting anomalous behavior in urban parking lots. It proposes a mobile application that allows users to collect data on location, time, and license plate details. The collected data is then analyzed using advanced time series analysis methods to identify anomalous behavior, such as unauthorized parking, irregular occupancy patterns, and violations of parking regulations. Real-world experiments on diverse parking lot datasets demonstrate the high accuracy of the proposed approach in detecting anomalies. These insights are valuable for predicting future parking demands, enabling parking administrators to efficiently allocate resources during peak hours and optimize space utilization. Additionally, the analysis can detect irregularities in parking patterns, promptly identifying unauthorized or abnormal parking and violations, such as parking the wrong type of vehicle or parking in restricted or reserved areas. This work advances the state-of-the-art in time series analysis for parking lot management, providing valuable insights for practitioners and researchers in the field. It also contributes to more efficient, data-driven, and proactive parking management strategies, leading to improved urban mobility and enhanced user satisfaction.