It is estimated that up to 30% of traffic in cities is due to drivers looking for parking. Research suggests that drivers spend an average of 6-14 minutes looking for an available space in London. This increases individuals stress levels as well as congestion and pollution. Intelligent transportation system wants to provide an effective way to address these challenges by making parking easier and less stressful. Parking Guidance Systems are an important component in Intelligent Transportation Systems (ITS), which reduces search time for parking by presenting drivers with dynamic information on parking. An accurate prediction and recommendation analytics algorithm is the key part of the parking guidance system, which combines real time cloud-based analytics and historical data trends, and can be integrated into a smart parking user App. The paper proposes a prediction algorithm based on Transient queuing theory and Laplace transform to predict parking occupancy or predict open parking locations. First, it will describe the Parking Guidance Systems, which integrates the front-end and back-end together. The core of the system is an analytics engine to predict parking availability based on historic and real time data feeds. The system combined different data sources and established a data engine to migrate them into a robust system. Second, the study use queuing methods to estimate demand for parking space by location, bay types, parking rules and time, or specifically, historical data on parking usage collected in different cities. Finally, the paper also discusses a routing strategy which gives the best route to check available bays and optimize travel time and customer preference. The agent based simulation model is developed to evaluate the system performance and robustness.