Freight Commodity Flow Modelling and Data Exchange Processing

We were data scientists responsible for commodity flow modelling, traffic data processing, and manipulating huge databases for a proprietary freight market intelligence and traffic data tool.  This database integrates information from various publicly-available and privately-collected data sources, and provides an overall picture of freight flows within the U.S., at a commodity and county or MSA level of detail.  It is a feeder database to many statewide freight plans.  To build or update this database, we started with Census data, local economic data, and public railroad waybill sample, and augmented the data with specific information gathered from a proprietary motor carrier data exchange program.  Where specific information is not available, the public data is disaggregated or synthesized using established optimization methods, allocation matrices, and gravity-attraction models. We were specifically responsible for the following market segments: agriculture, coal, minerals, air freight, carrier data exchange, and barge transload sectors.  We also contributed to the body of modelling knowledge by developing and/or refining new data collection and manipulation methodologies, resulting in continual improvement of the product.  This proprietary tool continues to be available on a subscription basis from its owner.

Note: Alex Lu performed this work as an employee of another firm.