Understanding the variations in travel by time of day is essential to predicting transportation system performance and air quality impacts of the transportation sector. As tour- and activity-based modeling procedures become more commonplace, the need for accurate time-of-day modeling procedures that are sensitive to changes in policies or travel conditions, such as congestion, that affect time-of-day choices, is clear. FHWA recently conducted a research project to develop new methods of modeling travel by time of day that are sensitive to these concerns. One of the products of this project is a time-of-day choice modeling procedure designed to be applied within tour- or activity-based travel modeling processes. This procedure was designed to be compatible with most existing tour- and activity-based models in the United States. The time-of-day choice model is estimated by using household activity survey data and transportation level-of-service information from the highway network. This procedure was tested by using the activity-based travel model for San Francisco County, California. The time-of-day modeling procedure is described, including the model estimation process, data requirements for estimation and application, and the results of the San Francisco test application. The test application included alternative scenarios designed to validate the model's sensitivity to land use, transit level of service, traffic congestion, and time-of-day pricing.