The role of simulation as part of a predictive tool is quickly growing as the capabilities of the available tools and hardware has allowed for faster computational times. These new applications of micro, meso and macro scopic simulations are being used to predict near future traffic conditions and evaluate in real time the impact of demand or capacity limiting events, such as road works and traffic accidents on the road network. One such implementation of near real time modeling is the San Diego I-15 Integrated Corridor Management System that uses model-based predictions as part of a state of the art Decision Support System. This system helps operators to be proactive rather than reactive to changes to their road network. Using Aimsun Online, this model uses both analytics and microsimulation to predict traffic conditions at five-minute intervals over the next 60 minutes and evaluate potential response plans. As with typical planning models, the operational model used within the system is initially based on a detail model validation, however unlike the traditional use, where the model is used to understand conditions in the far future (5,10, 20, 30 years ahead) this model has been calibrated and validated to understand the conditions 15, 30, 45 and 60 minutes ahead and is constantly updating in order to run 24/7 operations. Using historical data sets and live data feeds from external sources (detectors, transit automatic vehicle location, signals and others) each simulated run is automatically created with the matching real world status and checked against the real data for validity. Finally, unlike a planning model where the existing conditions are from a signal snap shot of the network, the online model needs a set of safeguards and tools to help track when significant roadway improvements are made or when developments are completed that dramatically change the traffic patterns in the area. This paper describes the validation steps taken while calibrating a model of this type and the ongoing validation, checks and maintenance to insure the relevance and accuracy of the model on a day to day basis.