Freight and Land Use Travel Demand Evaluation: Final Report

Section 3
Topic D: Public and Private Data Sources

Key initiatives under this topic include the following:

• Update the current listing of best data sources from NCFRP 25, Freight Trip Generation and Land Use, by building on the work done in SHRP2 C20 projects and filling gaps identified in that research.
• Develop a national strategic plan for freight data-sharing to leverage private sector decisionmaking resources and make these more readily available to the public sector.
• Facilitate understanding within the community of practice (including both public and private sector practitioners) of the increasingly disparate time horizons affecting goods movement decisionmaking. Explore how the private and public sectors respond to that demand with policies, pricing, and traditional and innovative approaches to increasing supply chain capacities.

Topic D also provides additional resources for consideration.

Current Listing of Best Data Sources

NCFRP Report 25 provides guidelines for sharing freight data, primarily between public and private freight stakeholders. The report recognizes the difficulties in obtaining data from private entities as well as the significant costs associated with data collection, and provides examples of how to overcome these barriers. Additionally, the report highlights 18 different public and commercial data sources that can be assessed by practitioners without restrictions (see Table 3). It should also be presented in the context of the value of blending public and private data sources as described elsewhere throughout Topic D.

Strategic Plan for Freight Shipping Data Development and Sharing

Data confidentiality is a concern voiced by shippers and carriers. Given small margins on typical freight transportation services, shippers and carriers are often cautious about giving away their competitive advantage if competing firms have access to their data. At the national level, regulations such as Title 13 of the U.S. Code directing private firm participation in the Economic Census helps ensure the relevance of the CFS. As data interests become more fine-grained, however, balancing private and public sector interests becomes more challenging. For example, the SHRP2 C20 project team working in the Albany, NY, region was able to collect data after developing a non-disclosure agreement (NDA) with each participating firm. The process of developing and executing the NDA took substantial time and effort before data collection could begin.

A national strategic plan for freight shipping data would help standardize the development of more localized yet transferable metadata. This effort would require a public-private partnership (at least in the informal sense of the term) to balance the costs (of both fiscal and proprietary natures) and benefits (in terms of smoother goods movement policy and project implementation). Conceptual elements could borrow from both institutional frameworks such as NAICS code evolution as well as industry self-promotion techniques, such as Leadership in Environmental and Energy Design certification, to find an appropriate balance between incentivizing and compelling private sector participation. This strategic plan would also need to anticipate and incorporate technological and societal changes that would keep the database sufficiently nimble over time, a subject explored further in topic matrix D.

Aligning Private Sector Shipping Needs with Public Sector Policies and Investments

Investment decisions in transportation infrastructure and services have long benefitted from robust data collection, compilation, and analysis. For the past several decades, this process has generally been led and managed by the public sector. Examples include demographic data such as those provided by the U.S. Census Bureau; travel flow information from traffic counts (both permanent stations and temporary project-level observations) and traveler surveys; and performance measures such as level of service and cost-benefit ratios. The public sector has also defined the rules and regulations for transportation system investments, ranging from National Environmental Policy Act (NEPA) impact analysis to best practices by agencies such as the American Association of State Highway and Transportation Officials, ITE, and TRB guidance. Major transportation system investments might have a useful life of several decades, so the process by which investment decisions are made has similarly valued analysis attributes such as authority, reliability, and transferability over attributes such as innovation and speed.

The information explosion discussed earlier in this report has shifted the data analytics paradigm. Recent advancements in transportation data collection have led to improved data quality, greater temporal coverage, wider geographical coverage, and differing population characteristics. New data sources, such as probe vehicles, GPS, crowd-sourced data (e.g., INRIX, HERE, Google, TomTom, Waze, etc.), Bluetooth, cellular data, and other emerging data sources could generate conventional performance measures needed for both traditional analytic methods as well as new measures for gauging factors such as reliability and resilience.

The information explosion has also illuminated the differences in evaluation timelines between transportation system providers and transportation system users. Other important factors to consider on this topic include:

Major transportation system infrastructure investments such as new roads, rail lines, and ports still have a useful life of 30 or 40 years and high capital costs that are supported in part by that lengthy amortization period.

Rolling stock such as trucks and railcars still have a useful life of 10 to 20 years, in part due to the certainty of the investment in the infrastructure they use.

Unlike infrastructure and rolling stock, the shelf life of data-intensive technologies driving market demand such as product innovations, supply chain management, product information distribution and customer feedback methods, and purchase transactions is continually shrinking and might best be measured in weeks or months rather than years.

The discrepancies between the relatively stable shelf life of the transportation system and rolling stock and that of the economic markets creates a challenge in aligning private shipper needs with public sector policies and investments. Market economies creating and using data are focused on near-term rates of return on investment; the state of the practice in the private sector today is almost certain to be obsolete in five to ten years.

Additional challenges associated with private sector data include cost, limited availability with strict use policies and NDAs, and privacy concerns. The public sector must also ensure that new and innovative data sources meet acceptable quality criteria; these include timelines, coverage, accuracy, continuity, and provenance. Despite the myriad challenges, transportation agencies negotiate with private vendors to obtain consistent, timely, quality data for planning purposes. Practitioners are bombarded with information; for example, the National Performance Management Research Data Set (NPMRDS) provides an important data resource and many State DOTs and MPOs purchase their own freight databases. However, in many cases these efforts are still insufficient to support decisionmaking on data spending.

NCHRP 49-14, Methods to Acquire Proprietary Data for Transportation Applications, will develop a synthesis of the benefits and limitations of non-traditional data for transportation applications. The project will identify available data sources at various levels (private and public sector), consider methods and tools such as data fusion techniques to apply innovative data sources to more traditional performance measure evaluation, and summarize how agencies might integrate new data sources to either complement or replace existing transportation data sets.

Truck Data Collection Survey Instruments

Additional materials for truck data collection include the development of automated data collection and interview techniques. These are described in more detail below.

Automated Vehicle Locator Resources
The Florida DOT SHRP2 C20 project on petroleum flows in southeast Florida included a summary of the strengths and weaknesses of vehicle detection technology characteristics associated with tanker truck identification. Table 4 provides excerpts from this summary to identify vehicle detection technology characteristics. A number of these methods could potentially be used to collect data on truck and freight movements.

Truck Data Collection Survey Techniques

The ITE Trip Generation Handbook is one of the foundational resources for this project. As described in the annotated bibliography, the ITE Trip Generation Handbook is a best practice that is periodically updated and adopted by ITE through a rigorous peer-review process. The third edition of the ITE Trip Generation Handbook covers a wide range of topics beyond goods movement. Thomas, Yarbrouh, Anderson, Harris, and Harrison (Transportation Research Record 2160, 2010, p. 163) recommended to ITE an application for a truck data collection survey. The Winston-Salem MPO SHRP2 C20 implementation assistance project in the Piedmont Triad applied this type of survey and provides a current example that would be appropriate for inclusion in the QRFM as a case study.

2017 CFS

The CFS is undertaken through a partnership between the U.S. Census Bureau and BTS. The CFS was initiated in 1993 and is conducted every five years (years ending in "2" and "7") as part of the Economic Census. The CFS produces data on the movement of goods in the United States. It provides information on commodities shipped, their value, weight, and mode of transportation, as well as the origin and destination of shipments of commodities from manufacturing, mining, wholesale, and selected retail and services establishments. The continued application of a steady approach provides both a snapshot of relatively current conditions as well as a baseline for longitudinal trendline analysis.

The CFS captures data on shipments originating from selected types of business establishments located in the 50 States and the District of Columbia. The establishments are asked to provide shipment information about a sample of their individual outbound shipments during a pre-specified, one-week period. For shipments that include more than one commodity, respondents are instructed to report the commodity that makes up the greatest percentage of the shipment's weight.

The current QRFM summarized the results of the 2002 CFS, noting that the coarse aggregation of the CFS database (the national data is aggregated into 114 zones) limits application CFS data. Since then, the CFS has been substantially disaggregated; the 2012 CFS included 3,143 zones. The 2017 CFS results are expected to be published in 2019. The content of the QRFM should be broadening to incorporate more explanatory materials. These materials would help decisionmakers and practitioners gain a more holistic perspective of the range of tools available.

Review of Freight Data Sets

Freight data collection is often fragmented and uncoordinated, and is often not well defined. Data standards can vary greatly across freight data sources, making it difficult to integrate available data. Further, Federal agencies have recently made national freight datasets available to the public; the data are typically not available on the sub-regional scale needed for effective local freight planning.

As part of the SHRP2 C20 program, the Capital District Transportation Committee (CDTC), the MPO for the greater Albany-Schenectady-Troy New York area, assembled a project team and developed a process to effectively collect, integrate, and maintain freight-related data from multiple sources, including innovative, easily obtainable private data and commonly used public databases. The project team identified existing freight data at the national, State, and local levels, and it designed and conducted data collection activities to obtain new freight data for the CDTC region. Figure 11 provides freight dataset characteristics from the Capital District study that facilitate data mining for local freight planning purposes.

The Capital District study focused on creating a dynamic freight database that would allow for aggregation of data at the local level. The study characterizes a comprehensive list of potential freight datasets, sources, processes for obtaining data, compatibility with research, level of disaggregation, advantages, and limitations.

The study included publicly available information as well as data from shippers and carriers in the region, as well as survey and in-depth interview responses. Together, this information helped CDTC understand freight shipment patterns in the region and provided insight into the decisionmaking processes by these businesses. A major positive outcome of this project was identifying the availability of processed data that will readily support any level of aggregation. For example, the project showed that the CFS and FAF can support national-level analyses; EZ-Pass or truck count data can model interstate flows; and freight trip generation, freight generation, and service trip models can support ZIP code-level analyses. In addition to obtaining and preparing the database, the project team calibrated models to estimate the freight flows (trips, generation, and services) at the ZIP code level.

You may need the Adobe® Reader® to view the PDFs on this page.