There is no strict priority order for data sources used for the FCPI project. The primary objective is to demonstrate a functional predictive AI architecture that successfully fuses a variety of multimodal public and private data sources to achieve a baseline prediction accuracy. Offerors should propose what they believe to be an optimal mix of datasets to support the intended use: “to enhance both public and private freight operational decision-making and to gain system-wide freight efficiencies.” For public data sources, the government anticipates provision of these for model training, calibration, and validation. For private-sector commercial data, offerors should not assume the government will automatically furnish private commercial data for Phase I efforts. One of the core Phase I deliverables is the Data Inventory and Integration Plan, which requires defining the specific access methods and availability for chosen data sources. Therefore, if an offeror’s solution explicitly relies on obtaining new, direct commercial licenses for proprietary private data, this should be included in the strategy and budget estimates.