Connected and automated vehicle (CAV) technologies attracted extensive attention in the past decade. As CAV brings convenience to travel, people's travel behaviors and patterns might change significantly. Existing models, however, cannot comprehensively evaluate the impacts on transportation systems. This study adopted an activity-based approach to evaluate the comprehensive CAV impacts on the transportation system in Southern California. A stated-preference survey was conducted, and captured people's behavior changes associated with CAV deployment. The model prediction demonstrated that the total trip number increased by 9%, with an 13% growth in total car-like mode travel distance. Among all trip purposes, work trips contributed to 49% of total trip number growth and 75% of the increased car-like mode travel distance. The advanced CAV technology alone wouldn’t directly benefit future transportation systems and it is still critical to have appropriate policy interventions in place.