Future freight and logistics survey: An integrated vehicle-and-shipment-tracking data collection method and a case study in the United States

High-quality freight data are essential for transport planning and decision-making, particularly when conducting freight analysis and modeling. Conventional freight surveys have limitations, such as low data collection efficiency and a lack of detailed shipment and vehicle activities. Moreover, few previous studies have developed an integrated survey methodology to collect high-quality freight data and conducted a practical pilot study to verify its feasibility. To address this research gap, this study provides a systematic approach to collect high-quality data and reduce the burden on both the surveyors and participants. Based on the Future Mobility Sensing (FMS) platform, the survey is implemented. It utilizes mobile sensing devices and wireless communication technologies to capture data on the movements and activities of vehicles and shipments. They are then interpreted into freight diaries using machine learning algorithms. A pilot study was conducted in the United States, where the vehicles and shipments of each participating establishment were continuously tracked for 1–3 weeks. This effort collected high-resolution GPS trajectories and identified 806 vehicle stops with detailed activity data captured at each stop. The 97.27 % (784 out of 806) detection success rate for vehicle stops and the 81.81 % (72 out of 88) success rate for generating shipment timelines have demonstrated the feasibility of this integrated approach. The novelty and key contributions of this study are threefold: (1) It proposes a fully integrated and feasible freight survey methodology that involves all key decision-makers (i.e., shipper, carrier, driver, and receiver) to collect comprehensive freight activity data that reveal actual behaviors. (2) The proposed methodology includes simultaneous tracking of shipments and vehicles, which supports the matching of these flows to reconstruct the complete transport process. The dual-tracking also enables cross-checking when shipment or vehicle data are missing. (3) It introduces a platform-based multi-party collaborative verification mechanism to support all decision-makers in verifying the collected data directly. This enhances both data accuracy and reliability. Overall, the framework involves multiple decision-makers and provides a holistic view of the entire freight process, offering a significant advancement over traditional freight surveys. Moreover, the comprehensive dataset collected through this integrated approach supports the model development, especially for activity-based models and agent-based models, which are essential for evaluating logistics performance and informing freight policy-making.