Freight Data Sharing Guidebook (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

13 Twenty-eight guidelines were developed based on identified barriers to effective freight data sharing and measures that have been taken to overcome the barriers. Table 3.1 shows the 28 guide- lines organized into six categories. In the following pages, the guidelines are defined under their cor- responding categories in more detail and, most importantly, illustrated with one or two examples from the various case studies that the research team analyzed. It is hoped that these examples, which constitute best practices, will be useful to public-sector freight planners, private-sector freight data providers, public and private freight partnership leaders, and freight data practitioners. The guidelines start at the initiation of a public sector project. Most public sector analyses of the transportation system, whether at the local or metropolitan area or at the national level including crossing international borders, involve the analysis of transportation data. The ques- tion for public sector analysts and their consultants is where to get the freight data they believe they need to support their studies. The guidelines include many examples of projects that had a need for freight data and found ways to overcome barriers to getting the data. Guideline 1 – Identify Sources of Freight Data via Literature Search and Review of Past Research Note that the first guideline is considered to be a cross-cutting one that applies to all freight data sharing projects. Fortunately, there is a large body of research that has been conducted and published that identifies freight data sources, particularly those that are publicly available and either have no cost or restric- tions or are available for purchase. This first guideline contains a list of commonly used and publicly available data sources. The key to this guideline is to do a literature search and take full advantage of past research. Not reinventing the wheel can save both time and money. Some of the past research studies are enumerated here and others are included in the references section of the contractor’s final report, which is available on the TRB website (www.TRB.org, search for NCFRP Report 25). Examples included with the guidelines include projects that built on the foundation of publicly available data and explain how they supplemented those data sources in conducting their projects. For example, NCFRP Report 10: Performance Measures for Freight Transportation (2011) notes many sources of freight data. The list of public and commercial data sources shown in Table 3.2 is based on a list from NCHRP Synthesis 410: Freight Transportation Surveys (2011) (Table 19, page 32). 3.1 Guidelines Related to Nonrestricted Data Two guidelines were identified that apply to the use of some of the public data sources listed or other data that might not require restriction. Table 3.3 shows those guidelines and examples that exhibited those guidelines. Leaders of example projects talked about difficulties in obtaining C h a p t e r 3 Freight Data Sharing Guidelines

14 Freight Data Sharing Guidebook Table 3.1. Summary of guidelines for freight data sharing. 1 2 3 agreements. 4 . 5 6 7 8 rly in the process as 9 blic agency policy approaches to support data . Identify sources of freight data via literature search and review of past research. Guidelines Related to Nonrestricted Data Use nonrestricted or open source data if available. Utilize nonintrusive technologies for data collection that do not require sharing Guidelines to Address Privacy Concerns If unrestricted data is not enough, be aware that privacy concerns must be addressed A nondisclosure agreement can be a good tool to support a data sharing arrangement. A stable contracting relationship with data providers can be very helpful in successful data sharing. A less formal agreement to maintain confidentiality of private sector data may be sufficient. Begin negotiations of disclosure and use restrictions on freight data as ea possible. Public agencies desiring to obtain data from private companies may need to research Freedom of Information Act (FOIA) laws. 10 Consider seeking enabling legislation and pu sharing and protect the data. Guidelines for Scrubbing or Restricting Access to Freight Data 11 Consider the use of software and database tools to protect and access freight data by removing private or competitive information. 12 Build access restrictions into the data set as an alternative to scrubbing Guidelines for Stakeholder Engagement 13 Place a high priority on coordination and devote the needed resources to extensive coordination with public and private stakeholders. 14 Consider the use of trusted third parties (associations, consultants, or academics) as intermediaries or data analysts. 15 Investigate possible partnerships with trade associations to facilitate data sharing. 16 Coordinate with local or regional agencies that may have closer relationships with data providers. 17 Consider gradual implementation of data acquisition coupled with coordination about successes. Guidelines for Articulating Benefits of Sharing 18 Define and articulate the benefits, goals, and purpose of data sharing to stakeholders. 19 Include a stipulation that data is for limited or for one-time use and cannot be used for any other purposes such as regulation. 20 Publicize the cooperation amongst project partners and seek to give the project visibility to stakeholders and the public. 21 Explain clearly to stakeholders that sharing of data will support improved freight infrastructure decisions that will benefit those stakeholders. 22 Add value to the data and make it available to all stakeholders. 23 Use technologies that are useful for other purposes. 24 Explore new market opportunities with potential data providers. Guidelines for Funding for Data Sharing and Projects 25 Attempt to include funding for research and data collection in public sector contracts. 26 Be sure to include funding to cover costs of data sharing and needed agreements to protect data. 27 Where appropriate, try to obtain joint public-private funding for projects. 28 Consider gathering data from volunteer stakeholder groups or roundtables.

Freight Data Sharing Guidelines 15 Table 3.2. Public and commercial data sources. Database Source and URL Airline Traffic, Airfare, and Airline On- time Data U.S. DOT Bureau of Transportation Statistics – BTS Publicly available at www.bts.gov Border Crossing Data U.S. DOT Bureau of Transportation Statistics – BTS Publicly available at www.bts.gov Commodity Flow Survey (CFS) U.S. DOT Bureau of Transportation Statistics – BTS Publicly available at www.bts.gov Freight Analysis Framework (FAF-3) U.S. DOT Federal Highway Administration Publicly available at www.ops.fhwa.dot.gov/freight/freight_analysis/faf/ faf3/netwkdbflow Industry Trade Data and Analysis U.S. Department of Commerce International Trade Administration publicly available at www.trade.gov/data.asp Intermodal Data and Statistics Intermodal Association of North America (IANA) - Multiple report and products – some for a fee. See www.intermodal.org/statistics_files/index.shtml Maritime Statistics Reports and Survey Series and Fleet Statistics U.S. DOT Maritime Administration.. publicly available at www.marad.dot.gov/library_landing_page/data_and_ statistics/Data_and_Statistics.htm National Roadside Survey Commercial Vehicles Surveys U.S. DOT/Federal Motor Carrier Safety Administration publicly available at www.fmcsa.dot.gov/documents/facts-research/CMV- Facts.pdf Port/Import/Export Reporting Service (PIERS) available for a fee from Journal of Commerce at http://www.piers.com/ Rail Waybill Sample Surface Transportation Board publicly available at www.stb.gov/stb/industry/econ_waybill.html Rail Industry Operating Statistics Association of American Railroads www.aar.org/StatisticsAndPublications.aspx - Multiple products – some publicly available and some only available to members Rail Industry Reference Files RAILINC Corporation multiple products, some publicly available and some available to subscribers only at www.railinc.com/ References and Files tab State of the Trucking Industry American Trucking Associations - Only available to the subscribers. See www.trucking.org/StateIndustry/Pages/default.aspx Ton Miles of Truck Shipments by State U.S. DOT Federal Highway Administration Publicly available at www.ops.fhwa.dot.gov/freight/freight_analysis/nat_ freight_stats/docs/09factsfigures/ Transborder Surface Freight Data U.S. DOT Bureau of Transportation Statistics – BTS Publicly available at www.bts.gov TRANSEARCH IHS Global Insight - Available for a fee at http://www.ihs.com/products/global- insight/industry-analysis/commerce- transport/database.aspx TranStats – The Intermodal Transportation Database U.S. DOT Bureau of Transportation Statistics – BTS Publicly available at www.transtats.bts.gov Waterborne Commerce of the U.S. U.S. Army Corps of Engineers - publicly available at www.ndc.iwr.usace.army.mil/wcsc/wcsc.htm Sources: NCHRP Synthesis 410: Freight Transportation Surveys, Transportation Research Board of the National Academies, Washington, D.C., 2011.

16 Freight Data Sharing Guidebook permission to use some datasets and the ease with which other data, without restrictions, could be used. Guideline 2 – Use Nonrestricted or Open Source Data if Available Most of the freight data collected by government agencies or trade and industry associations can be accessed without restrictions. For example, U.S. DOT Bureau of Transportation Statistics (BTS), as noted in Table 3.2, provides access to a wide range of aggregated freight data for differ- ent modes. In a similar manner, the European Commission’s statistical office (Eurostat) collects country-level transportation data that includes freight data. Other sources of freight data address more narrowly defined areas. The Washington State University (WSU) lock outage study used the data from the U.S. Army Corps of Engineers’ Waterborne Commerce Statistics Center. Some modes (e.g., railroads, trucking) and many industrial trade associations (e.g., wheat, soybeans) have web sites and reports with readily available freight data. Practitioners should look first at these studies and the sources they used. Lessons learned by others can make it easier to get the most from existing data. NCFRP Project 03, published as NCFRP Report 10: Freight Performance Measures for Freight Transportation (http://www.trb.org/Main/Blurbs/165398.aspx), discusses the need to develop national freight measures to gauge the performance of the freight system. The final report rec- ommends approaches, including a national freight report card. For the report card, the report suggests drawing on multiple unrestricted data sources, mostly from government agencies but also from the private sector or trade associations. The report has an illustrative list of data sources. Many of the sources listed in Guideline 1 above were identified in NCFRP Report 10. The Minnesota Department of Transportation sponsored a project called “Measurement Sources for Freight Performance Measures and Indicators” that identified many data sources including open source data as described under Example 2-1. Example 2-2 includes the use of open sources in the Columbia River Lock Outage study. Guideline 3 – Utilize Nonintrusive Technologies for Data Collection That do not Require Sharing Agreements If a freight project covers a metropolitan or regional area, it is likely that the data sources in Guidelines 1 and 2 will need to be supplemented by current, locally collected data. This guideline addresses data that can be collected easily and without the need for complicated agreements. Certain freight data collection technologies are nonintrusive by nature and do not require formal agreements. Radio frequency identification (RFID) tags, for instance, can be used to measure border crossing and wait times at international land ports of entry. The tags are already ubiqui- tous on trucks enrolled in the Free and Secure Trade (FAST) program, used for weigh-in-motion systems (e.g., PrePASS), and trucks equipped with toll transponders (e.g., EZ-Pass in a number Guideline Project Examples 2 Use nonrestricted or open source data if available • Minnesota Freight Performance Measures Project • Snake River Lock Outage Study 3 Utilize nonintrusive technologies for data collection that do not require sharing agreements • Southern Border Wait Time Project • Border Crossing Information System Table 3.3. Summary of guidelines for nonrestricted data.

Freight Data Sharing Guidelines 17 Example 2-2. Impacts of Columbia-Snake River Extended Lock Outage The WSU’s Freight Policy Transportation Institute completed a study that assessed the transportation and environmental impacts of extended lock outages on the Columbia-Snake River System. Information collected for this effort includes the volume and variety of commodities shipped up and down river. One important open source of data was the Lock Performance Monitoring System (LPMS) main- tained by the U.S. Army Corps of Engineers. As each barge passes through each lock, the content of the barge is reported to the lock operator for LPMS coding. The content of each barge is recorded in the North American Industry Classifica- tion System (NAICS). This unrestricted information is available through a central database: http://www.ndc.iwr.usace.army.mil//.htm. Another unrestricted data source used for this project was information from trade associations. One example of an association that provided information is U.S. Wheat Associates, which provides trade information (http://www.uswheat.org/.nsf/OpenPage). More details about the lock outage analysis project can be found at: http://www.fpti.wsu.edu/.htm. Example 2-1. Minnesota Freight Performance Measures Project To support Minnesota’s Freight Plan, the Minnesota DOT conducted a project that identified and recommended the best sources of information available for state level freight performance measures (FPM) and indicators. A significant aspect of the project is that it compiled, analyzed, and classified information sources for all modes. The recommended data was from a wide range of sources and most of these were open source and unrestricted. Sources identified included: • Federal Data Sources. CFS, Waybill, Waterway data from Army Corps, and Eco- nomic and Industry Surveys by Census. BTS and FHWA data (particularly FAF data). • State Sources. Past freight-related studies and statewide and district plans. Eco- nomic, demographic, establishment, export and import, and other information available from economic development departments and other Minnesota agen- cies. Data from commercial vendors. Some data from technology devices such as roadway loops. • Private Sources. IHS Global Insight’s TRANSEARCH for national and regional flows. Association of American Railroads data on various aspects dealing with rail freight. The IANA data on intermodal freight. PIERS and AAPA for water- way and port data. Annual logistics survey sponsored by the Council of Supply Chain Management Professionals provides good insight into the factors that are affecting performance of freight industry. The intent was to use the data sources to develop freight performance measures and indicators to evaluate the performance of Minnesota’s freight transportation system. The project was completed in July 2008. Details about this project can be found at: http://www.lrrb.org//200812.pdf.

18 Freight Data Sharing Guidebook of eastern states). Detection systems can be designed in such a way as to read the RFID chip’s ID number and a timestamp, but not record the specific truck/company moving the freight. Bluetooth readers detecting enabled cell phones or other mobile devices also have been used in this way. Although these arrangements do not require a data sharing agreement, it is important to coordinate with stakeholders including trucking companies to keep them informed about what is going on. If readers need to be installed on government property (such as a border crossing facility), appropriate permission is required to locate and connect the equipment. However, this can often be done informally through stakeholder meetings and interagency coordination. Extra time should be planned to allow for coordination with all of the agencies involved, particularly in a border environment where customs/security, facilities management, law enforcement, and transportation agencies (often from two countries) will be at the table. Example 3-1 describes a project in which RFID chips and toll tags were utilized to compute border crossing and wait times at U.S.-Mexico border crossings in Texas. Example 3-2 shows how Bluetooth technology has been used in a similar manner along the southern border (although the Bluetooth readers were detecting cars, the same method has been tested for trucks). 3.2 Guidelines to Address Privacy Concerns To really understand the movement of freight through a metropolitan area, freight ana- lysts need detailed freight data from transportation companies and shippers. These private firms are concerned that data about their operations might be used by their competitors to gain business advantage. While firms know they need to comply with regulatory require- ments involving data, they are sometimes reticent to share data with public agencies without protection of the details of their business and customers. Analysts and practitioners who want private data must take positive steps to address these concerns. The willingness of all Example 3-1. Southern Border Wait Time Measurement Projects FHWA sponsored a study of the use of RFID from existing passive tags to capture crossing and wait times of U.S. bound commercial vehicles on the U.S.-Mexican bor- der. Readers were installed at key spots on both sides of the border. FAST tags and others used for tolls in the area were read, so no additional tagging was needed. The RFID system captures the tag ID and timestamp, but not the truck’s ID. This assures anonymity and protects individual companies. No agreement with trucking companies was needed because the tags were already installed for other purposes. Permission to locate RFID readers was obtained through stakeholder meetings with U.S. Customs and Border Protection and the General Services Administration. Subsequent to the implementations in El Paso, similar measurement systems using RFID technology were implemented at other border crossings in Texas including Pharr-Reynosa International Bridge, World Trade Bridge, and Camino Colombia Bridge. A similar system is being implemented at the Veteran’s Memorial Bridge in Brownsville, Texas, and Mariposa Port of Entry in Arizona. More information about this project can be found at the Texas Transportation Institute web site: http://tti.tamu.edu/projects/project_details.htm?id=2497 and the FHWA publications web site: http:ops.fhwa.dot.gov/publications.

Freight Data Sharing Guidelines 19 parties to remove sensitive information from data to be exchanged or to restrict the uses that can be made of data can be important in gaining acceptance of data sharing. Table 3.4 below shows seven guidelines and some case study project examples that exhibited those guidelines. The following guidelines address privacy concerns in more detail. The critical nature of protect- ing privacy of data cannot be overemphasized. There are many examples of dealing with privacy Example 3-2. Border Crossing Information System Project The purpose of this project was to develop a prototype of a centralized repository of border crossing-related data and provide traveler information to the public to aid in their decisions about the time of day to make a border crossing, as well as provide performance-related data to stakeholder agencies. The prototype was developed by using the Paso Del Norte Regional Mobility Information System (PDN-RMIS) in El Paso as the platform. Border crossing-related archived data and pre-trip traveler information were integrated with the PDN-RMIS database and the pre-trip traveler information was added to the existing PDN-RMIS web site. Both Bluetooth and RFID data from readers installed near the border were used to compute travel times. In the prototype, data were extracted from Customs and other agency web sites. Since the data are public, there was no access agreement necessary. It was noted, however, that changes in agency web sites could not be accommodated without database management efforts within the Border Crossing Information System. In a fully operational system a formal data sharing agree- ment between agencies would be beneficial in terms of minimizing these data management requirements. Additional information is available at the TTI web site: http://tti.tamu.edu. Guideline Project Examples 4 If unrestricted data is not enough, be aware that privacy concerns must be addressed. • Cross Town Improvement Project (C-TIP) • Wireless Waterways Project 5 A nondisclosure agreement can be a good tool to support a data sharing arrangement. • Truck Data from GPS Vendors • Electronic Freight Management – Kansas City 6 A stable contracting relationship with data providers • can be very helpful in successful data sharing. Canadian GPS Projects 7 A less formal agreement to maintain confidentiality of private sector data may be sufficient. • Mississippi Study of Intermodal Technologies 8 Begin negotiations of disclosure and use restrictions on freight data as early in the process as possible. • Otay Mesa Border Delay Project • International Trade Data System 9 Public agencies desiring to obtain data from private companies may need to research Freedom of Information Act (FOIA) laws. • The FHWA and ATRI’s Freight Performance Measures initiative 10 Consider seeking enabling legislation and public agency policy approaches to support data sharing and protect the data. • Canada Freight Gateways and Corridors Project Table 3.4. Summary of guidelines for privacy concerns.

20 Freight Data Sharing Guidebook concerns including long term stable contracting relationships and also standard nondisclosure agreements or NDAs. Guideline 4 – If Unrestricted Data is not Enough, be Aware that Privacy Concerns Must be Addressed Since freight is usually carried by the private sector, public sector program use of private freight data depends, to some degree, on the public sector’s ability to collect private sector data. This is often a challenge since the private sector is reluctant to release data for a range of reasons including business and privacy concerns, lack of resources to make the data available, fear of government regulation, and a sense that there is little value to giving the public sector their data. Most likely, if freight data is not publicly available from one of the numerous data sources listed in Guideline 1, it will be from private sector sources and its use and access may be restricted. This critical need to protect and restrict the use of freight data is often due to privacy concerns: freight data can be business sensitive, proprietary, or reveal personal information. A July 2007 TRB workshop on freight data noted the following: Anecdotal evidence shows that the basis for industry’s reluctance to share this data derives primarily from, in order of importance: fears of civil litigation; competitive access; and regulatory impacts. With satisfactory resolution of these concerns, industry would be much more likely to participate in data- sharing partnerships with government and academia. The public sector must also demonstrate how the data can be collected and maintained in such a way that the private sector has a reasonable level of confidence that the confidentiality of their data is pro- tected. (Pages 41-42, “Meeting Freight Data Challenges Workshop,” July 2007.) Feedback received at the March 1, 2012 Freight Data Sharing workshop, as well as the research conducted on this project, has confirmed that there are downsides to data sharing for private companies. Similar to the noted challenges, the workshop identified privacy, resources, fear of regulation, or competitive concerns. It is therefore incumbent on public agencies requesting data to recognize these drawbacks and mitigate them where possible. The public sector can reduce these privacy concerns by aggregating any data output to obscure individual company’s information, signing legal documents such as nondisclosure agreements, promising not to use the data for regulatory actions, providing resources to help private com- panies prepare appropriate outputs of their data, and demonstrating how each company’s input will result in positive changes. There are many examples of projects that needed to address privacy concerns. These included the FHWA Performance Measures Project performed by the American Transportation Research Institute (ATRI), the Washington State and Southern California GPS projects, and the port analysis projects at the Ports of Los Angeles and Long Beach. Typically these privacy concerns were expressed by private sector partners who were asked to provide their freight data to the public sector. Addressing these concerns often resulted in the use of neutral or trusted third par- ties, development of legal agreements, or other mechanisms to protect privacy. The Cross-town Improvement Project in Kansas City described in Example 4-1 only made progress after the data from the private sector terminals was protected. Example 4-2 describes the Wireless Waterway project on the Ohio River. Guideline 5 – A Nondisclosure Agreement Can be a Good Tool to Support a Data Sharing Arrangement In the United States, a common tool to support data sharing is a nondisclosure agreement (NDA). An NDA is a legal contract or document between two or more parties that restricts any unauthorized disclosure of confidential data and can require reasonable measures to

Freight Data Sharing Guidelines 21 Example 4-1. Cross-Town Improvement Project (C-TIP) This project was a technology application designed to improve the efficiency of cross-town truck movements in Kansas City that connected railroad yards. The project attempted to maximize productive moves and minimize unproductive ones. It consisted of several components including real-time traffic monitoring/ dynamic route guidance for draymen; an open architecture exchange of load data and availability information between railroads, terminal operators, and trucking companies; and wireless information exchange for truckers regarding trip assign- ments, traffic congestion, trip status, and location. This required the collection of supply chain event data and tracking of truck movements between rail terminals. Some partners were reluctant to share proprietary data such as commodity or financial information but were willing to sharing data such as origin and destina- tion of the container to coordinate the moves between terminals. Some private sector partners, particularly the railroads, were willing to participate only after their freight data was protected by using a trusted third-party integration con- tractor. In a few cases, the data sharing was not possible or was limited because privacy concerns could not be adequately addressed. More information on this project can be found here: http://www.ctip-us.com. Example 4-2. Wireless Waterways Project The Port of Pittsburgh Commission (PPC) sponsored a series of interrelated stud- ies by Carnegie-Mellon University that culminated in the design of a network to allow key stakeholders (barge operators, the Corps of Engineers, and the U.S. Coast Guard, and other waterway interests) to replace outdated workflow pro- cesses such as faxes and phone calls with more efficient digital communication technologies. The Corps developed a series of digital communications-compatible tools to collect data from the barge operators. The Corps found that much of the data it wanted was already being collected by industry. The voyage, commod- ity, and vessel data submitted by industry is to be protected from disclosure that would reveal individual company operations. The system as envisioned would allow data owners (e.g., the barge lines) to decide what data becomes public and what is proprietary. It has become clear over the years that there are certain data that would be of interest to the public sector and which the barge industry is willing to share, but no one has ever asked them for it. The Wireless Waterways system would speed up the gathering of information now done manually, but it would allow for the gathering of so much more data that is not currently even collected. It is anticipated that there would remain an owner of the data, pro- tected by firewalls. The PPC is providing the seed money for funding for the initial test bed and demonstration network on the Ohio River. The concept was com- pleted in 2006 and funding is needed to implement the results of the test bed. More information about this project can be found at http://www.port.pittsburgh. pa.us/home/index.asp?page=175.

22 Freight Data Sharing Guidebook prevent any such disclosure. The NDA may require that the organizations receiving the data limit access of the data only to employees having a need to know in connection with the data. NDAs also may require the agency originally owning the data to identify what information needs to be considered confidential and to clearly define what constitutes autho- rized usage. Specific elements that can be included in the NDA to protect freight data include: • Information about the parties involved; • A concise definition as to what data needs to be protected; • Any up-front data cleansing requirements by the suppliers; • Any fees related to the data sharing; • When the contracts and agreements should be updated; • Language about how long the non-disclosure agreement should last; and • Legal information about resolving disputes and terminating the relationship. NDAs are considered legal documents and it may be wise to have an attorney review the docu- ment. This will increase the time and resources required. Often these NDA arrangements can use a fairly standardized format and many larger institutions have NDAs in preset formats. Appen- dix B contains a sample NDA from a university transportation research group and Appendix C is a consulting firm NDA. A search on the internet will find a number of additional sample NDA agreements. Each project is different and may require a unique NDA; however these two appen- dices are examples of data sharing NDAs that have been used successfully in the past. Project examples that involved NDAs include the FHWA/ATRI and Washington State performance measures projects (described in Example 5-1), the Southern California Truck data analysis proj- ect, the Electronic Freight Management case study in Kansas City (Example 5-2), the port truck- ing movement study at Los Angeles/Long Beach, and the International Trade Data System in the Federal government. Example 5-1. Truck Data from GPS Vendors Both the ATRI FPMs project for FHWA and Washington state truck GPS case stud- ies use data obtained from private sector GPS fleet management vendors. The use of vendors circumvented direct concerns about the individual company’s business sensitive information, but the vendor still required privacy protection in the form of nondisclosure agreements. The data feeds between the vendors and ATRI and Washington State were set up only after NDAs were developed to help protect the data. The NDAs stipulated or involved both legal punitive actions as well as technical approaches such as suppression of individual trucker’s names and allow- ing only the release of aggregate data. ATRI has data sharing agreements with recipients of their services, but the core agreements between ATRI and its data providers are considered confidential. ATRI executes nondisclosure agreements and contracts with each provider of data as well as with each recipient of the GPS data. The contracts and agreements are updated annually and as projects and uses are defined. These relationships have evolved over time based on trust and sensitivity of use of data. Information on the ATRI effort can be found here: http://www.atri-online.org. Detail on the Washington State GPS project can be found here: http://www.wsdot.wa.gov/freight/.

Freight Data Sharing Guidelines 23 Guideline 6 – A Stable Contracting Relationship with Data Providers Can be Very Helpful in Successful Data Sharing The FPMs GPS truck data project and the Canada Borders project both had longer term contractual arrangements that supported data sharing. See Example 6-1. The ATRI, the not-for-profit research arm of the American Trucking Associations, has a long-term Example 5-2. Electronic Freight Management – Kansas City The purpose of the Electronic Freight Management (EFM) project is to conduct case studies to document the benefits of companies using Web services and elec- tronic data exchange technologies in different supply chain scenarios. The goal is to improve supply chain efficiency and reduce supply chain costs for shippers and carriers through a series of case study tests. In Kansas City, the project involved a distributor receiving ocean containers by rail. Supply chain event data (rail depar- ture, in transit, arrival, customs clearance, delivery) was captured and shared. Partners who participated in EFM were concerned with protection of their exist- ing information systems from unauthorized access. This was mitigated through data security layers and digital certificates for transactions between authorized partners. Confidentiality concerns related to partner data were mitigated by memoranda of understanding (MOUs) to protect from outside access. In Kansas City, there was a single multi-party nondisclosure agreement for the 2–3 month deployment test. Tests were conducted to assure accuracy of automated data acquisition and to analyze the benefits of automation. Researchers found that each partner’s needs and view of proprietary data were different. This required considerable coordination and effort by the study contractor. There are two web sites that provide additional information about EFM case studies: http://www.efm. us.com and http://www/efm-saic.com. Example 6-1. Canadian GPS Projects Both Transport Canada and the Ontario Ministry of Transport (MOT) have developed long-term contracts with one of the major GPS providers, Turnpike Global Technologies, Inc (TGT) for equipping trucks, collecting data, scrubbing it of IDs, and then sharing the data with the appropriate government agen- cies. Two projects included in the research involved TGT data, the Canadian Gateways and Borders project and the Canada Border Wait Time project. These projects included acquiring existing GPS data that TGT has collected from its extensive network of existing transportation industry clients. From the public agency perspective, the contractual relationship with TGT shifts the primary burden of working out multiple data protection agreements to TGT. The stable contractual relationship with a data provider is important in obtaining freight movement data. The Transport Canada web site has more information about this project at http://www.tc.gc.ca/eng/policy/acg-acgc-menu_gateways- 1961.htm.

24 Freight Data Sharing Guidebook contract with U.S. DOT/FHWA. ATRI then contracts with several GPS providers, collects data, scrubs it, and then uses it to support U.S. DOT objectives. Through its long-term contract with U.S. DOT, ATRI acts as an intermediary and combines together GPS data from multiple GPS vendors in building its GPS database. Since the trucking industry is usually not in the data sharing business, and as a research organization supported by the trucking industry, ATRI views itself as a trusted third party with trucking industry relation- ships that help to acquire data. ATRI executes a nondisclosure agreement and contract with each provider of data. The contracts and agreements are updated annually and as projects and uses are defined. The University of Washington and WSU as well as other academics such as the Texas Transportation Institute at Texas A&M have successful long-term contracts as well. Guideline 7 – A Less Formal Agreement to Maintain Confidentiality of Private Sector Data may be Sufficient Where privacy is less of a concern, formal contracts or NDA’s may not be needed. For proj- ects where freight survey data is collected using interviews, a data sharing agreement was often as simple as a statement that the responses and data would be held confidential. An example of such agreements is a mail-out survey as part of the Mississippi ports technology study (see Example 7-1). Guideline 8 – Begin Negotiations of Disclosure and Use Restrictions on Freight Data as Early in the Process as Possible Negotiating nondisclosure agreements or restrictions in the use of data is very time con- suming, so those negotiations should begin as soon as possible. In addition to addressing data privacy concerns, the negotiation process can be needed to determine prices and set up funding arrangements. Acquiring data for the Otay Mesa border crossing study (see Example 8-1 below), the Southern border Wait Time, and Washington State GPS truck data project all required a negotiation process. Example 8-2 describes a data sharing project within the federal government that required detailed approvals and restrictions between users and the controlling agency, in this case Customs and Border Protection. It should be noted that such negotiations also consume corporate legal resources. Example 7-1. Mississippi Study of Intermodal Technologies The purpose of this 2001 project was to assess the use, adoption, benefits, and impacts of intermodal information technologies on intermodal ports and termi- nals serving agribusiness firms in Mississippi. Surveys were developed and sent to port and terminal operators in the state. Survey participants were asked to share data both to help understand port industry in Mississippi and to help them better understand their use of technology in context of the global economy. As public agencies also interested in public relations, port officials and terminal operators were willing to share information. The survey participants were told their replies would be held in strict confidence. The survey cover letter told participants that the study promised to hold the data confidential and that it would not be possible to extract individual business information. See the following web site for more information: http://ncit.msstate.edu/PDF/mso2B0.pdf.

Freight Data Sharing Guidelines 25 Example 8-1. Otay Mesa Border Delay Project The project began in October 2007 to assess GPS and license plate recognition technology for the measurement of travel times for trucks through the Otay Mesa international border crossing from Mexico into the United States. The project collected GPS truck movement data across the border from January 2009 to March 2010 for participating trucks crossing the border and allowed analysis of border crossing times and delays against standards. Prior to the data collection, the data provider participated in a lengthy negotiating process before the carriers finally agreed to grant access to their data. Before collection, stakeholder sessions with transportation industry interests, border agencies, local planning agencies, and the public offered insights into the chal- lenges faced by users and administrators at the border. These sessions allowed for a comprehensive understanding of the conditions as they existed at the border and what might be done to improve them. This process showed project participants that care is needed in defining specifically what data is to be col- lected. They also found that more specificity of data needs to be provided in a contract, including spelling out in more detail the data processing involved. The conclusion was that the level of time and effort necessary to execute agree- ments with the carriers indicates that this should be factored in whenever GPS fleet data is sought. More details are included in the final report at http://ops. fhwa.dot.gov/publications/fhwahop10051/fhwahop10051.pdf. Example 8-2. International Trade Data System ITDS is a U.S. Customs and Border Protection (CBP) system for sharing trade data with other government agencies. The purpose of the project is to develop a single window for import processing for federal government agencies that regulate the import or safety of goods. The single window helps avoid proliferation of parallel import reporting systems. ITDS provides controlled access to Automated Customs Environment (ACE) data. ITDS provides an interface and appropriate data sharing agreements for other government agencies to access ACE data from CBP. CBP negotiates MOUs with other government agencies to obtain ACE customs data and in some cases to acquire additional information on import shipments. The MOU covers how each subordinate agency within the particular government agency will access data. Individual users are required to complete nondisclosure agreements. The agreements and the interface in ITDS control access to the industry-sensitive data and provide a single data gathering process with industry. Negotiating concepts of operations, memoranda of understanding, and non- disclosure agreements with all of the 47 various agencies is a challenge. Standard- ization of data elements and operating procedures for other agencies to use ACE take considerable coordination and time to complete. See http://www.itds.gov for more information.

26 Freight Data Sharing Guidebook Guideline 9 – Public Agencies Desiring to Obtain Data from Private Companies May Need to Research FOIA Laws For governmental agencies, privacy protection may be complicated by open information laws that allow individuals to request information from governmental agencies. For example, the 1966 FOIA is a federal law that gives individuals access to any U.S. government agency records unless the release fits within nine exemptions (or release is prohibited by law). The exemptions include confidential business data and personal privacy that can support freight data sharing. In spite of these exemptions, individuals in the private sector can have the perception that public agencies are unable to protect data due to FOIA. Most states also have their own version of freedom of information laws that are known as open government, open meeting, or sunshine laws. A number of national organizations catalog and provide information about each state’s laws. One such source is http://nfoic.org/state-freedom- information-laws. As with the federal laws, state laws typically have some exemptions related to protecting privacy that may allow freight data sharing. As Example 9-1 shows, innovative approaches may be needed to exercise these exemptions and protect privacy concerns and meet the intent of FOIA without divulging private data. Guideline 10 – Consider Seeking Enabling Legislation and Public Agency Policy Approaches to Support Data Sharing and Protect the Data There are a number of legal and policy protection approaches that can address these privacy concerns. At the highest level it can involve national laws that facilitate and protect data sharing. The Canadian Transportation Gateways and Corridors Project (Example 10-1) benefited from a national nondisclosure law, known as the Access to Information Act, that standardizes the protec- tion of data. This act provides access to data held by federal agencies, but also explicitly guarantees the protection of commercially sensitive information. The freight data provided to the Eurostat is based on legislative mandates that are required as part of being a European Union member as well as based on voluntary agreements. Example 9-1. The FHWA/ATRI Freight Performance Measures Initiative The purpose of this FHWA-funded project is to develop and test a national system for monitoring freight performance on the nation’s highways (http://www.atri- online.org). The program is led by ATRI who collects GPS truck position records data and scrubs them to remove individual companies’ identities. The actual data is obtained from a variety of sources including fleets, GPS vendors, and tele- communications companies. Users of the data obtain a license from ATRI for data products and services. The actual data is owned by the providers, who govern whether and how it can be shared. This approach means that FOIA and state sunshine laws can’t be used to access the raw data. ATRI acquires raw data, then licenses out rights to use it on a royalty-free basis. They can stipulate that it is a one-time use that can’t be used for any other purposes. These requirements help to ensure continued participation of carriers. More information is available at http://www.atri-online.org.

Freight Data Sharing Guidelines 27 3.3 Guidelines for Scrubbing or Restricting Access to Freight Data In a 2009-10 FHWA project entitled Freight Data Sharing Compendium, the primary barrier to freight data sharing identified was the possible disclosure of individual shipment or company data. The primary mitigation for this disclosure problem is aggregating or scrubbing data to remove individuality. While there are certainly nondisclosure and other agreements that govern data sharing, the Compendium project found that aggregating data was the practice that most protected data and yet allowed important research to go on. Table 3.5 shows two guidelines and the case study projects that exhibited those guidelines. Data users (agencies) can help mitigate these concerns by engaging in best practices, such as scrubbing sensitive data, signing nondisclosure agreements, providing funds to cover the costs of supplying the data, and involving a trusted third party to act as a data repository. Example 10-1. Canada Freight Gateways and Corridors Project Canada’s Strategic Gateways and Trade Corridors program was developed to ad- vance the multimodal integration of that country’s major transportation systems. An important part of the program is the development of a range of national gate- way performance measures. Developing measures for each geographic area has involved the collection of multimodal freight data from the private sector freight carriers. One reason this effort could collect private freight data was a national nondisclosure agreement supported by a Canadian law, the Access to Informa- tion Act, and a strong Transport Canada policy of data protection (supported by a visible history of protecting data). Like the American Freedom of Information laws, the Access to Information Act provides the public access to data from federal institutions. However, this act also legally protects commercially sensitive data collected by public agencies such as found in the situation for the Gateways and Corridors project. Canadian Federal agencies are restricted from releasing any sta- tistics that relate to any identifiable business without the previous consent of that business. Private sector freight organizations can be referred to this act and this increases their willingness to share freight data with the Canadian governmental agencies. More information on the act is at: http://laws-lois.justice.gc.ca/eng/acts/ A-1/index.html. More information on the gateway and corridor project is found here: http://www.tc.gc.ca/eng/policy/acg-acgc-menu_gateways-1961.htm. Guideline Project Examples 11 Consider the use of software and database tools to protect and access freight data by removing private or competitive information. • Modeling Freight in Alabama • Southern California Association of Governments Heavy Duty Truck Model 12 Build access restrictions into the data set as an alternative to scrubbing. • Lock Operations Management Application (LOMA) Table 3.5. Summary of guidelines for scrubbing data.

28 Freight Data Sharing Guidebook Guideline 11 – Consider the Use of Software and Database Tools to Protect and Access Freight Data by Removing Private or Competitive Information A common approach to overcoming data privacy concerns is to scrub the data of all com- pany, shipment, and/or operator or driver information prior to delivering it to a public agency. This has been the approach for several truck GPS tracking projects including the ATRI and Washington State performance measures projects, the Southern California Truck modeling project, and the Canada Gateways project. GPS data can be automatically cleansed of specific company or vehicle identifying information while still retaining truck position records and timestamp information which is valuable for freight planning. Often, it is easier to do this through a third party data provider, such as a firm that provides GPS devices and tracking ser- vices to logistics companies on a contract basis. Assuming the data owner is willing to release the information, this relieves them of having to devote resources to scrubbing it beforehand. Normally, negotiations about data release occur between the data owner (e.g., a logistics firm) and its GPS services provider. After securing permission to release the data, the provider will scrub it according to the stipulations set forth in its contract with the owner and release it to the public agency. As mentioned earlier, ATRI performs a scrubbing role with GPS data that is subsequently used in several projects including the FPMs project for FHWA, the Southern California Association of Governments project (see Example 11-1), and the Minnesota DOT Performance Measures study. A variety of database and software tools can be used to facilitate the data scrubbing and pro- tection. The simplest approach may be just to strip a database of selected columns or variables with identifiers. Alternatively, data hashing algorithms allow for specific information (such as company ID) to be transposed (hashed) into a new code. The new code remains within the data- Example 11-1. Southern California Association of Governments Heavy Duty Truck Model In its Comprehensive Regional Goods Movement Plan and Implementation Strat- egy, the Southern California Association of Governments (SCAG) used GPS data from existing commercial GPS truck tracking/operations vendors during the period from October 2009 to July 2010 to support the development of a heavy duty truck model. Historical GPS data from fleets already deployed in the six-county SCAG region were purchased from three established GPS vendors to support the data collection effort. Trip data that has been stripped of company ID is made avail- able to SCAG. The GPS data collection vendors were concerned with the risk of the data being used to specifically identify one of their customers (e.g., a specific trucking fleet); this concern was mitigated through sanitizing the data before it was provided to the public sector. Because the data obtained from the vendors was found to be inadequate, SCAG’s contractor purchased a license for GPS truck move- ment data products and services from ATRI. This data was already scrubbed, and after processing, was used to update the SCAG model. The difficulty associated with recruiting fleets for the GPS data collection effort may show that it is more viable to purchase data from secondary sources as opposed to recruiting fleets to participate in data collection. The reluctance of fleet managers to participate mainly revolved around the economic climate, privacy concerns, or a general distrust of government. Additional information can be found at: http://www.scag.ca.gov/goodsmove/.

Freight Data Sharing Guidelines 29 set and is internally consistent to allow for functions such as geographic tracking but reduces the ability to identify specific businesses. This approach was used by the vendors that supplied GPS data to the Washington State Freight Performance Measures project. Another more complicated technical approach is to data mask using statistical disclosure limitation tools such as used for U.S. Census data and other Federal agencies (http://www.fcsm.gov/working-papers/totalreport. pdf). Since much freight data is geographical, it also is possible to use GIS tools to filter data to just specific network segments, zones or jurisdictions relevant to a project. For example Inrix, a private sector company that sells truck performance data derived from GPS devices, provides roadway segment-level travel information. The use of segment-level information removes any individual truck data and any origin and destination-based travel patterns. Finally, database software can be set so users can only view and output aggregated data. Software and database approaches can be used to extract useful information from freight data that is aggregated or protected. The Alabama freight modeling study (see Example 11-2) developed a framework and database to combine aggregate regional Federal data with project- collected local data to develop usable local freight flow information that was input into freight models. Guideline 12 – Build Access Restrictions into the Data Set as an Alternative to Scrubbing Data scrubbing can be a labor intensive process that can add significant cost to a freight data collection effort. One way to mitigate this issue is to build suitable access restrictions into the freight data as part of the sharing arrangement. It is possible to develop automated processes which aggregate freight data to a level sufficient to protect privacy but still useful for planning efforts. Similarly, company or shipment-identifying information can be hashed or scrambled prior to delivery to a government agency (this is the approach used in the Washington State GPS project discussed in Example 5-1). Indeed the sheer size of such data- bases will oftentimes require automated data processing protocols. Database software also can be set up to allow different types of access to freight data based on different users’ autho- rization levels. Example 12-1 shows how carrier identifiers were deliberately left out of the Lock Opera- tions Management Application (LOMA), a system developed by the Army Corps of Engineers Example 11-2. Modeling Freight in Alabama The University of Alabama in Huntsville obtained and analyzed readily available federal data such as from the U.S. DOT’s Freight Analysis Framework, and used surveys to collect local shipping data. A methodology and database were devel- oped to combine these two levels of freight data for use in forecasting freight demand arising from the household sector. This combined information was used to analyze the local area through trips and to develop a freight O/D matrix for modeling freight flows in Alabama. This ability to combine aggregated freight data with specifically collected local data allowed the City of Mobile MPO to pro- duce an intelligent estimate of freight movement and helped to validate trans- portation models. See the following web site for more information: http://ntl.bts. gov/lib/32000/32100/32101/Forecasting_final_demand__pass_through_freight_ Final_09_14_0.pdf.

30 Freight Data Sharing Guidebook and its partners to collect and disseminate pertinent inland waterway operational data to lock operators and barge captains. Additional data streams may become available in future itera- tions of the system, but the initial deployment is already providing information valuable to lock operators. When developing an information reporting/dissemination system involving many public and private actors, it can help to have an institutional arrangement in place to help deal with the confidentiality issues that are likely to crop up. The LOMA project is leveraging other federal initiatives aimed at harmonizing data collection and reporting efforts across agencies. 3.4 Guidelines for Stakeholder Engagement It is incumbent upon project leaders, particularly public sector participants who desire data from other entities, to coordinate with everyone involved in a project. Coordination committees and numerous stakeholder meetings are important to the ultimate success of a project. Coordi- nation is needed to achieve data sharing and to accomplish large public-private projects. Failure to coordinate with all stakeholders can lead to failure of a project or a much longer and more expensive project or data sharing approval process. Five coordination guidelines are shown in Table 3.6 along with examples for each. Example 12-1. LOMA The purpose of LOMA is to provide the inland waterway navigation community (including lock operators, barge lines, government agencies, and the navigation industry) with improved situational awareness through the collection, integration, and dissemination of existing data streams as well as new ones. Overall the effort seeks to automate much of the manual data collection which is currently required by lock operators. The system leverages existing Automatic Identification System (AIS) equipment required by the Coast Guard on certain commercial vessels. This provides an opportunity to employ the technology for the broader use and exchange of important navigation data. In this initial deployment, only vessel location and speed data, plus weather information gathered from public sources, is available. Vessel identification infor- mation was deliberately left out of the initial phase due to privacy concerns, how- ever future phases may include this and other data points such as commodities carried/weight if confidentiality issues can be adequately addressed. Additional data elements such as lock gate settings and delay times, navigation advisories, tracks taken, and river stages and hazards (e.g., debris) are planned as the system is refined in the future. The Corps of Engineers is working with government and industry partners to address privacy and use concerns to enable further expansion of LOMA. The Fed- eral Initiative for Navigation Data Enhancement (FINDE) is an intra-governmental effort to harmonize navigation data collection and sharing. Under FINDE, the Federal-Industry Logistics Standardization (FILS) effort works with industry to agree on common data reporting standards and formats. These have proven to be good forums for addressing these types of issues. Articles that describe LOMA in more detail can be found at the Corps of Engineers web site: http://chl.erdc.usace. army.mil/.

Freight Data Sharing Guidelines 31 Guideline 13 – Place a High Priority on Coordination and Devote the Needed Resources to Extensive Coordination with Public and Private Stakeholders The more parties there are in a project, the more difficult it is to work through all of the coordination issues. This is particularly true of projects that involve international borders with multiple federal, state/provincial, and local governments as well as private firms in both Guideline Project Examples 13 Place a high priority on coordination and devote the needed resources to extensive coordination with public and private stakeholders. Electronic Freight Management Data Exchange 14 Consider the use of trusted third parties (consultants or academics) as intermediaries or data analysts. ATRI Freight Performance Measures Project 15 Investigate possible partnerships with trade associations to facilitate data sharing. Performance Measures for Freight Transportation 16 Coordinate with local or regional agencies that may have closer relationships with data providers. Washington E-Seal Border Crossing Project 17 Consider gradual implementation of data acquisition coupled with coordination about successes. Importer Security Filing (10+2) Project • • • • • Table 3.6. Summary of guidelines for stakeholder engagement. Example 13-1. Electronic Freight Management Supply Chain Data Exchange The goal of FHWA’s Electronic Freight Management program was to improve supply chain efficiency and reduce supply chain costs for shippers and carriers. The Kansas City EFM deployment test applied Web and Internet-based data exchange technolo- gies to a wholesale supplier that used data and reports about supply chain move- ments to better manage its operation. Supply chain partners included shippers and consignees (the principal supply chain owner); transportation providers including third party logistics providers, rail carriers, and local trucking companies; and a cus- toms broker. An integration contractor implemented the technology and conducted the test for FHWA. This partnership model was used in a previous Columbus test and with several subsequent supply chain case studies, but the partners were different for each case study. Determining what data each partner would provide and how it processed and transmitted the data required significant coordination amongst FHWA, the study contractor, the supply chain owner, and the various partners. Kan- sas City SmartPort, a not-for-profit corporation that facilitates transportation in the Kansas City area, was a participant in the promotion of the project and assisted in coordination. Confidentiality of partner data was mitigated by MOUs to protect data from unauthorized outside access. Each partner’s needs and view of proprietary data were different. This required considerable coordination and effort by the study contractor. Reports of results from additional case studies are or will be available as they are completed through FHWA and its case study contractors. Identification and calculation of benefits were useful parts of the reports that were shared and coor- dinated with partners. There are two web sites that provide additional information about EFM case studies: http://www.efm.us.com and http://www/efm-saic.com.

32 Freight Data Sharing Guidebook countries. Thus, stakeholder coordination was an important element in a number of successful freight data sharing efforts including several border crossing travel time studies and the CREATE rail improvement public-private partnership in Chicago. For the border crossing projects, coordination was more complicated because of the various government organizations at the borders (two border enforcement agencies, two sets of facilities managers, two state/provinces, at least two local gov- ernments, two federal governments). Projects around terminals that included terminal operators, shippers, and the carriers that served them involved coordination with diverse public and private interests. The FHWA-sponsored EFM supply chain improvement projects were good examples of such coordination (Example 13-1). Some projects had success with formal public-private coor- dinating committees and meetings with stakeholders. Another aspect of freight data sharing that appeared in several studies was that local or personal contacts, as well as personal appearances by project leaders at freight stakeholder meetings, enhanced survey and interview response rates. Guideline 14 – Consider the Use of Trusted Third Parties (Consultants or Academics) as Intermediaries or Data Analysts Experience with several projects has shown that personal relationships and trust between the public agency desiring the data and the private firms providing the data are important. Specifi- cally, some projects involved performing contractors who were trusted by both public and pri- vate sector partners. Examples include ATRI and its relationship with the trucking industry (see Example 14-1), several university-led projects including the Washington State GPS Performance Measures Program, Alabama freight study, and Texas border crossing projects, and the Cross- Town Improvement Project, where the integration contractor was well known to the railroads involved and had railroad experience. Guideline 15 – Investigate Possible Partnerships with Trade Associations to Facilitate Data Sharing Trade associations can help foster relationships as well as facilitate data sharing. As advocacy groups, trade associations typically understand how infrastructure policy decisions impact their industry and can therefore see the potential value in sharing data. They are also likely to Example 14-1. ATRI Freight Performance Measures Project As previously discussed, the ATRI conducted this project for DOT/FHWA and developed and tested a national system for monitoring freight performance on the nation’s highways (http://www.atri-online.org). As the research arm of the American Trucking Associations, ATRI was considered a trusted third party for near real time data collection. ATRI has a data sharing agreement with recipients of the services, but the core agreements between ATRI and its data providers are considered confidential. The trucking industry is usually not in the data sharing business, but program participants believe that it is important to have a trusted third party in this arrangement since the data owners need to be sure the in- formation won’t be used in ways that could harm them. ATRI has a multi-year government contract with FHWA because the trucking industry has indicated the results are useful and of value. ATRI’s relationship with the trucking industry allowed data to be collected and cleansed. Oftentimes, once participants know who ATRI is, they are willing to help.

Freight Data Sharing Guidelines 33 provide access to appropriate owners of freight data. For example, in the Minnesota Performance Measures study, the researchers obtained data dealing with rail freight from the Association of American Railroads, data on intermodal freight from the IANA, and waterway and port data on intermodal freight from both the PIERS and AAPA. Example 15-1 describes some of the data sharing involved in freight performance measures. In the study of the Snake River Lock Outage, a Technical Advisory Committee (TAC) included staff from the American Trucking Associations, and the Association of American Railroads. A lesson learned from that project was that dealing with trade associations can be more useful than dealing with individual businesses. Similarly, the Goods Movement Roundtable in the MetroLinx study in Greater Toronto and Hamilton included private sector industry associations as members. Guideline 16 – Coordinate with Local or Regional Agencies that May Have Closer Relationships with Data Providers Metropolitan Planning Organizations (MPOs) or other local/regional agencies can some- times collect data more easily than higher level agencies. These agencies may have a specific, focused working relationship with data providers. Example 16-1 involved collecting border data Example 15-1. Performance Measures for Freight Transportation NCFRP Project 03, which culminated in NCFRP Report 10, developed a comprehen- sive set of performance measures for the nation’s freight transportation system. Measures are presented as a Freight System Report Card, which has three levels of increasingly detailed information to serve the needs of a wide variety of stake- holders. The Report Card includes 29 performance measures in six categories, and reflects different levels of geographic detail from the local to the global per- spective. The proposed freight report card would draw on multiple data sources, mostly from government agencies but also from the private sector or trade asso- ciations. See the following for more information: http://apps.trb.org/cmsfeed/ TRBNetProjectDisplay.asp?ProjectID=1575. Example 16-1. Washington E-Seal Border Crossing Project The project involved tracking of containers with electronic seals moving both north and south between Seattle and Vancouver, British Columbia. Pilot tests involved in-bond containers sealed in the Puget Sound Ports and read at the U.S.-Canadian border. The objective was to enhance security and reduce transit time at the border crossing. The funding for this project was from the U.S. Depart- ment of Transportation, Washington State Department of Transportation, U.S. Department of Homeland Security, the State of Washington, Whatcom Council of Governments (the local MPO) and their IMTC. The IMTC organization was helpful in dealing with stakeholders and had data sharing agreements and relationships from previous tests. Researchers in the project found that data sharing depends on personal relationships, especially with the enforcement agencies and that it is easier to maintain trust at the local level. Additional information can be found at: http://www.wcog.org/imtc.

34 Freight Data Sharing Guidebook for the electronic seals project from the Whatcom Council of Governments’ (a Washington State MPO) through their International Mobility and Trade Committee (IMTC) rather than CBP. That MPO had a long term and strong relationship with border enforcement agencies operat- ing in their jurisdiction. This provided access to data that was not readily available to other public agencies. Guideline 17 – Consider Gradual Implementation of Data Acquisition Coupled with Coordination About Successes Sometimes starting small is helpful, particularly if it is coupled with continuing coordination and feedback with data providers and, as appropriate, the public. Not every project lends itself to an evolutionary approach, but there are good examples of projects that succeeded because they incrementally expanded the scope after early phases. A good example is the ATRI performance measures project which started as an analysis of five corridors, with data collection from compa- nies specifically related to operations on those highway corridors. Subsequently, the relationships, data agreements, documentation of results, and overall coordination allowed ATRI to expand the scope of its data collection to become nationwide and involve trucks operating across the border with Canada. In the 10+2 project, CBP stretched out the implementation period so that the coor- dination with shippers and carriers could be completed and so that individual companies would have the time to comply with the data submission requirements (see Example 17-1). In both projects, these confidence-building measures allowed involved parties to become comfortable with the data sharing before making big commitments. 3.5 Guidelines for Articulating Benefits of Sharing It is important for project proponents to be able to explain to the public, private sector par- ticipants, and other stakeholders how they will benefit from the conduct of a project. Articulat- ing benefits is an important part of coordination of a project. This is needed at the beginning when the participants are seeking approval of a project and very often needed throughout the conduct of a project. Sometimes publishing analyses of the expected costs and benefits of a Example 17-1. Importer Security Filing (1012) Project The purpose of the Importer Security Filing (ISF), commonly known as the 10+2 Project, is to obtain import container manifest information prior to vessel departure from a foreign port and require automated submittal of that information to CBP prior to departure of the vessel from the foreign port. After the introduction of ISF in early 2009, CBP had a period of gradual enforcement and extensive public relations and coordination with importers and carriers. CBP stated at the time that they really wanted the data and not penalties, and that they did not want to disrupt the flow of legitimate cargo into the United States. CBP performed extensive outreach and phased in the enforcement of the requirements to ease the ability of importers to comply. The preparation, involvement of the trade, and the gradual enforcement were important in gaining compliance and acceptance of the program by all parties involved. The CBP web site has additional information: http://cbp.gov/xp/cgov/trade/cargo_ security/carriers/security_filing/.

Freight Data Sharing Guidelines 35 project helps to assure its success. There are seven guidelines that deal with benefits of sharing, shown in Table 3.7, along with examples from case studies. The case studies and the workshop conducted under this project both identified the impor- tance of articulating benefits. Guideline 18 – Define and Articulate the Benefits, Goals, and Purpose of Data Sharing to Stakeholders Almost all successful projects include analysis at the beginning to carefully publicize potential benefits of a project. Particularly in data sharing, it is crucial that project proponents articulate benefits to private sector participants who might not otherwise be willing to share data. Often it is important to stipulate that the data will not be used for regulatory enforcement. The key in this area is communications with project participants and with the public. Example 18-1 describes efforts in the Cross-Town Improvement Project (C-TIP) in Kansas City. Example 18-2 explains a slightly different situation at the Department of Homeland Security where, for mat- ters of national security, they want to require transportation companies and shippers to submit data; the DHS effort to articulate its goals is described. The coordination and outreach by cus- toms agencies helped in getting the import community ready to submit the data. Coordination among public and private partners and with foreign ports and multiple countries is challenging but crucial to successful improvements such as the Secure Freight Initiative (SFI) foreign port scanning project. Guideline 19 – Include a Stipulation that Data is for One Time Use and Cannot be Used for Any Other Purposes Such as Regulation Private sector data providers were more likely to share their data if they could be assured that there are limitations on the way data is used. Most importantly, private sector firms were Guideline Project Examples 18 Define and articulate the benefits, goals, and purpose of data sharing to stakeholders. C-TIP Project 10+2 Project 19 Include a stipulation that data is for one-time use and cannot be used for any other purposes such as regulation. Washington State FPM 20 Publicize the cooperation amongst project partners and seek to give the project visibility to stakeholders and the public. EPA SmartWay Freight for a Day Study 21 Explain clearly to stakeholders that sharing of data will support improved freight infrastructure decisions that will benefit those stakeholders. Detroit Windsor Bridge Double-stack Clearance Improvement Project 22 Add value to the data and make it available to all stakeholders. Washington E-Seal Border Crossing Project Border Crossing Information System Project 23 Use technologies that are useful for other purposes. Otay Mesa-Tijuana Border Travel Time Measurement Project 24 Explore new market opportunities with potential data providers. Washington State and Southern California GPS Projects Table 3.7. Summary of guidelines for articulating benefits.

36 Freight Data Sharing Guidebook Example 18-1. C-TIP The C-TIP Intermodal Transfer Project is a technology application, supported by U.S. DOT/FHWA, designed to improve the efficiency of cross-town dray move- ments between railroads by maximizing productive moves and minimizing unpro- ductive ones (e.g., bobtails). For the railroads involved, the C-TIP contractor found that as long as they could show the benefits, the railroads would usually agree to participate. The primary selling point from the railroads’ perspective was saving money on cross-town drayage rates through a more rational system of coordinat- ing and dispatching moves. The railroads did recognize that the main immediate benefit would accrue to the dray companies, but that this could translate into lower rates for the railroads over the longer term. As long as participation did not cost the railroads too much, and would result in reasonable benefits to the Kansas City region, they were usually willing to share. Railroads wanted to be good cor- porate citizens. More information on this project can be found here: http://www. ctip-us.com. Example 18-2. Importer Security Filing (ISF) (1012) Project The purpose of the 10+2 Project is to obtain import container manifest informa- tion prior to vessel departure from a foreign port and require automated submit- tal of that information to CBP prior to departure of the vessel from the foreign port. The 10 importer data elements and two ocean carrier data files allow CBP to identify potentially high-risk cargo through the identification of actual cargo movements and, at the same time, expedite the processing of lawful international trade by identifying low-risk shipments early in the supply chain. As noted earlier, CBP conducted extensive public relations and coordination with importers and carriers. Extensive publicity and coordination occurred to determine the impacts on importers and to facilitate the ISF data being provided to CBP correctly and in a timely way. Detailed data requirements were issued to avoid confusion and make it easier for importers to provide the required data. The ISF program helps CBP meet the congressional requirement to provide advanced data to U.S. ports and reduce the risk of terrorism via import ocean container. The coordination and outreach by CBP helped in getting the import community ready to submit the data. 10+2 provides ocean carriers greater confidence in the security of the ship- ment they are transporting, and increases the likelihood of an uninterrupted and secure flow of commerce. The CBP web site has additional information: http://cbp. gov/xp/cgov/trade/cargo_security/carriers/security_filing/. concerned about additional uses by the government beyond the purpose of the project at ques- tion. In particular, they were quite concerned about using data against them, for example, in the regulatory process. Therefore, it is important for the public agencies to put in writing the intended use of the data and an agreement that data will not be used for other purposes. The example below describes techniques that were used when obtaining private sector truck data for the Washington State FPM project (Example 19-1).

Freight Data Sharing Guidelines 37 Example 19-1. Washington State FPMs The state-funded Washington State GPS FPMs Project used data from commercial GPS devices in trucks to develop a statewide freight performance measure program. The project involves ongoing GPS-based probe truck movement data collected for the Puget Sound area since 2008 and for all of Washington State since 2010. The GPS data are used to support a statewide freight performance measure program that includes locating and quantifying truck roadway bottlenecks. There is a monthly fee for the acquisition of the GPS data from the GPS vendors. Concern by individual companies of release of their data was mitigated by aggregating data and removing individual company ID. In addition, the public agencies assured the GPS vendors that the data would not be used for regulatory purposes and that the agencies would safeguard the data while making effective and positive use of the data. More infor- mation can be found at the University of Washington’s transportation research web site: http://www.depts.washington.edu/trac. Guideline 20 – Publicize the Cooperation Among Project Partners and Seek to Give the Project Visibility to Stakeholders and the Public Most freight data projects involve significant publicity of the goals, objectives, and benefits of the project in general and of the particular data sharing. It takes considerable time and effort to perform the coordination and articulate the benefits. If providers understand what public uses of data are planned, it may help with sharing of data. This was true in several state studies including the WSU lock outage study, the Minnesota DOT and Alabama Freight Study. In some cases, participants benefitted from favorable publicity of being involved in a project. This even included peer pressure as playing a role in convincing other similar partners to participate in a project. The EPA SmartWay partnership is discussed in Example 20-1 to show how visibility of a project and participation benefits individual companies in the program. Example 20-2 describes the Freight for a Day study in Philadelphia that gave companies an opportunity to show off their freight efficiencies. In addition, the CREATE partnership in Chicago was among the projects that provided positive publicity that is valued by the participants. The FPMs projects for FHWA and the State of Washington clearly stated that the data collected would support freight system improvements. Guideline 21 – Explain Clearly to Stakeholders that Sharing of Data Will Support Improved Freight Infrastructure Decisions that Will Benefit Those Stakeholders There are numerous projects in which this approach works with agencies that have infra- structure responsibilities. These projects included the Washington State Freight Performance Measures project (for roadway improvement projects), the Mississippi Study of Intermodal Information Technologies (terminals and intermodal technology investments), the CREATE rail infrastructure improvement program in Chicago, and the Metrolinx Greater Toronto and Hamilton Area Urban Freight Study (road and rail infrastructure). Example 21-1 describes an international bridge crossing project in Detroit that has involved articulation of benefits of the project. Example 21-2 is the Double-stack Clearance rail improvement in Philadelphia.

Example 20-1. EPA SmartWay The U.S. Environmental Protection Agency’s (EPA) SmartWay partnership program began in the early 2000s to encourage and recognize energy conservation in freight transportation. The primary focus is on transportation companies, particularly truck- ing companies, railroads, and those companies that handle intermodal freight which voluntarily agree to implement energy improvements. EPA encourages companies to use the SmartWay logo and the companies’ participation in the partnership in advertising to let suppliers, customers, and the public know about benefits accruing from the energy improvements. Freight data is not shared in the project, but com- panies enter into partnership with EPA to voluntarily reduce energy consumption and emissions. Participating companies benefit from the positive publicity and use of the SmartWay logo in advertising. Data about partners and the energy conservation efforts they have undertaken are shared with EPA and often publicized. Companies apply to become partners in SmartWay and download a model which is used to document emissions reductions. Firms enjoy cost savings, public/peer recognition, and environmental achievement benefits. EPA commits to promote company par- ticipation in the program by posting partner names on the SmartWay web site and in related materials. As the program expanded, partners already enrolled publicized their participation through trade conferences, meetings with other companies in the industry, and marketing/promotional materials. This increased awareness of the pro- gram within the industry; as more and more firms became familiar with SmartWay, more started to sign up. The program continues and plans to expand to other areas of freight transportation including short sea shipping. More information can be found at the SmartWay web site: http://www.epa.gov/smartwaylogistics/index.htm. Example 20-2. Freight for a Day Study As part of its regional planning responsibilities, the Delaware Valley Regional Plan- ning Commission (DVRPC) promotes freight transportation efficiency improvements and economic development throughout the region. The Freight for a Day study was to be a freight scan for all modes operating within the two-state region around Phil- adelphia and was intended to provide a picture of the extent of freight operations and of the economic impact that freight transportation has on the region. The study involved one day of data collection (September 20, 2006) from numerous public and private partners with accompanying field visits to ports and terminals in the Phila- delphia area to observe freight movements into and out of Delaware Valley freight facilities. The results of all of the data collection and observation were pulled together in a document that provided a scan of freight activity at various points within the study region. There were no formal agreements for the data that was collected. Through the personal contact by DVRPC, the goals and objectives of the study and the data desired from each party were communicated. Since the study was intended to publicize the role of freight in the region through one time data collection, there were no issues of sharing use of the data. Field visits were useful as reminders to participants and as opportunities for participants to show off their operations. The report “Freight for a Day, September 20, 2006 An Elementary Guide to Understanding Cargo Shipments in the Delaware Valley,” can be found at the DVRPC web site http://www.dvrpc.org.

Freight Data Sharing Guidelines 39 Example 21-1. Detroit Windsor Bridge The Detroit-Windsor New International Trade Crossing, formerly known as the Detroit River International Crossing (DRIC) Project, began in 2000 as a partner- ship among U.S. DOT/FHWA, Transport Canada, the Michigan Department of Transportation, and the Ontario Ministry of Transport. The primary purpose of the proposed new international bridge crossing is to improve transportation in the Detroit-Windsor corridor and to reduce the projected economic impact of increas- ing freight delays at the border. The new crossing has significant economic effects in both Michigan and surrounding states and in Ontario. The top auto makers, many other businesses, the Ohio legislature, and most area Chambers of Commerce sup- port the new bridge project. There have been useful and well-publicized studies of the economic impacts of the international crossings on the respective economies of Michigan and Ontario. The intent of the four public agency partners noted above is to award a contract for the new bridge as a public-private partnership in which the contractor finances and builds the publicly owned bridge and receives toll revenue to cover the cost of the financing. During 2012, an interlocal agreement was signed between the state of Michigan and the province of Ontario. A Michigan ballot ini- tiative that would have essentially delayed or stopped the new bridge was defeated in November 2012. Political arguments and potential lawsuits continue. So while the project is an interesting example of public-private efforts to build infrastructure, it also shows that having well-executed economic analyses to justify infrastructure projects is important, but not sufficient. The politics surrounding the alternatives for the new crossing have been difficult to date and continue to threaten the future completion of the project. There are numerous news articles about the bridge pro- posal and its history. These include http://www.freep.com which published a special bridge issue in April 2011, detfreepressbridgeissue.pdf and a pro-bridge web site, http://www.buildthebridgenow.com. Example 21-2. Double-Stack Clearance Improvement Project The Double-Stack Clearance Improvement Project, which began in 2009, is a public- private partnership for CSX rail physical plant improvements needed to allow double-stack container cars to safely pass through existing rail lines in Philadelphia. The effort involves the CSX railroad, the state of Pennsylvania, the city of Philadel- phia, and other proponents in the Philadelphia area and will be completed in 2013. The improvements, two thirds funded by public agencies, involve reconstruction of tracks and/or bridges at 16 crossings in the Philadelphia area. The improvement avoids circuitous routing of 37 miles and cuts transit time by 5 hours. The reason for the cooperative working relationship is that the state of Pennsylvania and busi- nesses that operate there are likely to benefit from double-stack improvement, since the more efficient container operations result in fewer trucks on area roads and bridges. The DVRPC has helped local officials plan and advocate for improvements and has periodically published information about the project to keep the public informed. See the DVRPC web site for more information: http://www.dvrpc.org.

40 Freight Data Sharing Guidebook Guideline 22 – Add Value to the Data and Make it Available to All Stakeholders If partners see value in the data products offered by agencies and are offered access to the infor- mation resources that their data helps to generate, they may be more willing to provide data. This is especially true for public-public data sharing, i.e., the sharing of information between government entities. Although formal agreements are not always needed for this sort of sharing since the data is already in the public realm, agencies would have to devote staff time to providing and maintaining data feeds to other agencies for the purpose of assessing transportation system performance. Some public data contains private industry data that is considered confidential, so sharing and confiden- tiality agreements are necessary sometimes. The commitment to provide data to another agency is more likely to be forthcoming if the receiving agency adds value to the data and makes it available in a useful format for the providing agency. Example 22-1 discusses public sharing of container tracking data at borders. Example 22-2 describes a system for consolidating and publishing public border crossing and traffic information on a web site for public users. Guideline 23 – Use Technologies that are Useful for Other Purposes It is important that participating companies perceive a benefit for themselves in sharing data. Technologies such as GPS devices are useful for public agency measurement of transportation system performance, but can also provide useful data for the company itself. Negotiations to secure this type of data will normally be more involved since the data is proprietary and could divulge sensitive business information. Agencies should carefully define the data that they need, and how it will be processed, to reassure private firms that their data will be protected. Informa- tion sharing agreements need to consider alternative uses of the data to ensure that the maxi- mum value can be obtained. Although data obtained from a limited set of probe vehicles will prove useful for measuring freight performance, significant penetration of the chosen technol- ogy in the region of interest would be required to develop truly representative results. Example 23-1 details a project where the FHWA purchased GPS data from a third party ven- dor with carrier consent to analyze border travel times at the Otay Mesa-Tijuana port of entry. Example 22-1. Washington E-Seal Border Crossing Project This project involved the tracking of containers with electronic seals on shipments moving north and south between Seattle and Vancouver, BC. Pilot tests involved in-bond containers sealed in the Puget Sound Ports and read at the border. The objective was to enhance security and reduce transit time at the border crossing by pre-clearing trucks prior to their arrival at the border. E-seal data was combined with truck transponder data from other pilot tests to improve border crossing efficiency and enhance operations along the I-5 Corridor. Carriers and other supply chain part- ners who process shipments across the border provided shipping documents and sup- ply chain event data. Information provided included intermodal movement data for containers moving by truck either through the corridor or to and from regional ports. Since, in many cases, the benefits of data sharing were made obvious, the project partners were supportive. The project was security-oriented but this also resulted in a process that appealed to the private sector partners. The border wait time metrics generated by the project were made available to participants. Addi- tional information can be found at: http://www.wcog.org/imtc.

Freight Data Sharing Guidelines 41 Example 22-2. Border Crossing Information System Project The purpose of the Border Crossing Information System (BCIS) project was to develop a prototype of a centralized repository of border crossing-related data and provide traveler information to the public to aid in their decisions about the time of day to make a border crossing, as well as provide performance-related data to stakeholder agencies. BCIS consolidates existing data from multiple public sources and presents it on a web site accessible by the public. Data are grabbed from other systems and stored in a relational database for display on web-based maps. The data are used for public advisories/information about border crossings, delays, and spe- cial circumstances at borders. Since the data is public, no access agreement was necessary. Prior to the development of the BCIS, there was no system to extract data from existing systems to present archived border data to users of a public web site for travelers and users of the border crossing. Rather, each agency’s data was used only by that agency. It was found that adding value to data provided by stakeholder agencies, especially public agencies, is the best approach a bor- der crossing information system can use to guarantee continuous support from the agencies. Additional information is available at the TTI web site: http://tti. tamu.edu Example 23-1. Otay Mesa-Tijuana Border Travel Time Measurement Project The purpose of this project was to assess the effectiveness of a technology for automated capture of travel time for vehicles crossing the border by collecting and analyzing one year’s worth of travel time data. GPS data from a third party provider was selected for a one year test to record travel times through the bor- der zone at Otay Mesa, California. GPS was chosen in part because it is a reliable, proven technology that is trusted by users. GPS data from two participating car- riers was used to measure transit time across the U.S.-Mexico border and analyze differences between FAST and normal lane crossings. Although the negotiations to obtain the GPS data were lengthy, ultimately the carriers became more receptive as they recognized that the devices are useful for other functions as well. GPS and related fleet management applications can be useful for asset tracking and other functions such as fuel consumption monitoring and fleet maintenance. GPS data recorded in the project could be used not only for travel time calculation, but also for the identification of origins and destinations and route selection. By working directly with carriers while agreeing to protect sen- sitive information from distribution, the data provider was able to structure its data agreements to allow for the re-use of information for multiple purposes without incurring additional cost. More details are included in the final report at: http://ops. fhwa.dot.gov/publications/fhwahop10051/fhwahop10051.pdf.

42 Freight Data Sharing Guidebook Guideline 24 – Explore New Market Opportunities with Potential Data Providers Historically, agencies have had to make do with inadequate public data or engage in expensive data collection efforts to accomplish freight planning goals, even though it is widely known that superior data exists but is controlled by private firms. In some cases, however, third party data pro- viders may find that there is an untapped market for anonymous freight movement data that public sector agencies can use for transportation planning and modeling purposes. GPS data in particular is being used in this way to help agencies better understand truck movements and bottlenecks in their regions. Although the same privacy and use concerns discussed in other guidelines must also be addressed for GPS data, once this has been done it can lead to an ongoing data relationship with participating fleets and vendors. Example 24-1 discusses two similar GPS projects, both of which successfully used purchased GPS data for freight performance measurement/modeling. The cost of purchasing freight data, especially on an ongoing basis, is frequently a concern for planning agencies. This is particularly true in smaller regions which may not have the same resources as larger ones. This is one reason why such projects have heretofore been limited to one-time efforts or expensive projects in larger regions. However, in the case of GPS, it should be noted that increasing read rates on GPS devices, combined with growing fleet penetration, will likely lead to reduced costs and better quality data in the future. Example 24-1. Washington State and Southern California GPS Projects This example involves two similar projects since the first one in some ways led to the second. In 2008, Washington State began using data from commercial GPS devices in trucks to develop a freight performance measurement program. The effort was funded by the State Legislature and the University of Washington, with additional sponsorship from the Washington Trucking Associations and the Wash- ington State DOT. Data collection was coordinated by researchers at the University of Washington. Having attempted unsuccessfully to recruit multiple truck fleets to provide data, the researchers turned to a commercial GPS vendor. After lengthy negotiations with the vendor, researchers gained access to aggregated/anonymous truck movement information. They used the data to map truck bottlenecks and construction impacts on truck movements in the Puget Sound region. Once the carrier privacy concerns were addressed, the GPS vendor realized that this could be a new revenue source for the firm. See http://www.depts.washington.edu/trac for more information. In 2009 and 2010, a similar truck data collection effort was undertaken on behalf of the SCAG for the purpose of updating their heavy duty truck model. The decision to use this approach was based in part on the successful experience in Washington State, and indeed one of the same GPS vendors was used for the Southern California effort. SCAG purchased truck GPS positional records from three different vendors, each representing different sectors of the regional truck fleet. This work provided a far richer data set at lower cost than could be obtained through traditional trip diary approaches. Moreover, the project further raised the profile of the public sector market with GPS vendors, who now recognize public agencies as potential new clients for their data. Additional information can be found at: http://www.scag.ca.gov/goodsmove/.

Freight Data Sharing Guidelines 43 3.6 Guidelines for Funding for Data Sharing and Projects Most freight data projects originate because of a public and sometimes public-private interest in furthering a transportation objective. Federal, state, or regional/local planning or transportation agencies conduct projects that look at the performance of the transportation system, address the impacts of congestion, and analyze the costs and benefits of proposed infrastructure improvements. Such projects may be conducted by in-house transportation staff at the various agencies, but more often the projects are carried by contractors working for the agencies. Thus, an important, if not essential, aspect of projects involving freight data is funding to conduct the research and to facilitate sharing of data. Four guidelines related to funding are shown in Table 3.8 with examples for each. Making sure that the data sharing efforts are adequately funded is important. The most successful projects are those where participants are reimbursed for their costs and efforts and where public and private money are brought to bear in accomplishing results. The following four guidelines are provided to improve the effectiveness of funds and are especially relevant as funding becomes tighter. Guideline 25 – Attempt to Include Funding for Research and Data Collection in Public Sector Contracts An important part of public infrastructure programs is adequate planning at the beginning of the project. Several of the border crossing projects both on the U.S.-Canadian border and the U.S.-Mexican border were funded to improve traffic flow and reduce delays at the border. Electronic Freight Management was a series of research projects sponsored by U.S.DOT to demonstrate improvements in automation of supply chain data. There have been numerous projects involving the capture, aggregation, and analysis of GPS data and most were federal or state initiatives. All of these public contracts involve stakeholder groups in the local areas and require adequate resources in the contract to coordinate with the various stakeholders and put in place agreements that help assure participation. Government funding from the contracts helps offset costs and encouraged participation in the projects. Public sector proj- ects should include funds to support the private sector activities involved with participation as well as data collection. This helps assure success of the project. Example 25-1 discusses the FHWA GPS project, Freight Performance Measures, which featured a long-term contract- ing partnership between FHWA and the non-profit trucking industry research firm ATRI. Guidelines Project Examples 25 Attempt to include funding for research and data collection in public sector contracts. • FHWA-ATRI Freight Performance Measures Project • Border Crossing Wait Time Projects 26 Be sure to include funding to cover costs of data sharing and needed agreements to protect data. • Canadian Gateways • The CREATE Program 27 Where appropriate, try to obtain joint public- private funding for projects. • Double-stack Clearance Improvement Project • Truck Turn Time Study 28 Consider gathering data from volunteer stakeholder groups or roundtables. • Minnesota FPM • GHTA Urban Freight Study Table 3.8. Summary of guidelines for funding data sharing.

44 Freight Data Sharing Guidebook Example 25-2 explains in more detail some of the contracts involved with several of the border crossing projects. Guideline 26 – Be Sure to Include Funding to Cover Costs of Data Sharing and Needed Agreements to Protect Data If it costs data providers to share data and they do not otherwise perceive benefits, reimburse- ment sometimes helps. Thus, when defining a contract, be sure to include funding related to data acquisition and sharing. As shown in Example 26-1, the Canadian Gateway project provided a Example 25-1. Freight Performance Measures The long term FHWA contract with American Transportation Research Institute (ATRI), the not-for-profit research arm of the American Trucking Associations, to obtain GPS data and analyze freight movements has been discussed in earlier examples. ATRI executes an annually updated nondisclosure agreement and con- tract with each provider of data and has expanded the number of participants in both the U.S. and Canada. ATRI’s contract with the data providers and software that it uses to scrub the data and remove individual trucking company identifica- tion data have been important in getting the various parties to provide the de- sired data. The truck travel time data collected by ATRI is being used to calibrate network assignment models and to understand activity by time of day. A freight performance measurement web site looks at average truck operating speeds on various U.S. highway networks. The multi-year contract with FHWA is viewed by both ATRI and FHWA as a partnership that has been crucial to obtaining the data and developing freight performance measures. More information is available at http://www.atri-online.org. Example 25-2. Border Crossing Wait Time Projects The U.S.DOT FHWA and Transport Canada jointly sponsored a wait time technol- ogy evaluation in July 2010 at Peace Bridge in Buffalo and Pacific Highway in Blaine, WA. The evaluation involved vehicle identification technologies such as RFID, Bluetooth, License Plate Recognition, and the measurement or calculation of accurate border wait times. Separately, there have been pilot projects at El Paso, TX (RFID), and Otay Mesa CA (GPS), funded by the U.S. DOT/FHWA government contracts. Additional information about border crossings can be found at the FHWA web site http://www.ops.fhwa.dot.gov. Transport Canada sponsored the Canada Border Wait Time project and signed a MOU with several provinces to set up an Ontario-Quebec Continental Gateway and Trade Corridor. The project involves a GPS-based border travel (transit) time system. After a successful proof of concept, Transport Canada awarded a contract to a third party GPS provider. More information can be found at: http://ebtc. info/2009_presentations/11_tardif_transit_times.pdf. These projects were successful because they provided public contract funds to project participants to pay for their efforts in participating.

Freight Data Sharing Guidelines 45 stipend to trucking companies to help with their costs of providing data. Example 26-2 below discusses the CREATE program in Chicago that used TIGER funds to support data sharing, coordination, and infrastructure improvements. Guideline 27 – Where Appropriate, Try to Obtain Joint Public-Private Funding for Projects Particularly where the public benefits from the work, rail improvement projects often involve both public and private funding. This motivates both the public agencies and the private com- panies to participate. Although it was not a data sharing project, the Double-Stack Clearance Improvement Project, described in Example 27-1, was successful because of the involvement and funding from both a state DOT and a railroad. The railroads also contribute substantial funding to the CREATE program in Chicago. Example 26-1. Canadian Gateways Transport Canada is the sponsor of the project to develop indicators measuring the reliability or fluidity of multimodal freight movements in Canada and across the U.S. border. Data is collected from private partners who are transportation providers, primarily trucking companies, but also from railroads when rail ship- ments are involved. Transport Canada has letter agreements with carriers to provide automated data from their systems. In addition, Transport Canada has a long-term contract with the GPS data provider. Transport Canada was also willing to assist financially any of the participating companies if they had to make minor adjustments in their information systems in order to provide the data. In addi- tion they assured the companies that there would be no ongoing, recurring cost. http://www.tc.gc.ca/eng/policy/acg-acgc-menu_gateways-1961.htm. Example 26-2. The CREATE Program The Chicago Region Environmental and Transportation Efficiency (CREATE) Program is a partnership to improve rail transport efficiency and reduce traffic congestion in the Chicago area via infrastructure and operational projects. The partnership includes local and state DOTs, the regional commuter rail system (Metra), Amtrak, six Class I railroads, and two switching railroads. CREATE receives U.S. DOT funding, state and local DOT funding, and significant funding from the freight railroads. There was extensive coordination with citizen groups to help mitigate problems. The various public agencies involved included the state of Illinois and local community governments including Chicago, who were involved because of the potential public benefits of rail improvements. Private businesses including shippers and railroad companies were also in the partner- ship. A key advantage of the partnership was in obtaining funds including earmarked federal funds as well as state, local, and private funds for the rail improvements. For more detail, see the Case Studies Appendix C of NCFRP Report 2: Institutional Arrangements for Freight Transportation Systems (June 2009), pages C-33 through C-46.

46 Freight Data Sharing Guidebook The Turn Time Study at the Port of Long Beach, described in Example 27-2, was funded by several different private groups and the port operators; it used data collection equipment and data collected in an earlier project that received federal funding. Projects to improve air quality in Southern California almost always include co-funding by private sector partners and/or other government agencies. These can be in the form of monetary or in-kind contributions. Example 27-1. Double-Stack Clearance Improvement Project The Double-Stack Clearance Improvement Project, which began in 2009, is a public-private partnership for CSX rail physical plant improvements needed to allow double-stack container cars to safely pass through existing rail lines in Phila- delphia. The effort involves the CSX railroad, the state of Pennsylvania, the city of Philadelphia, and other proponents in the Philadelphia area. The improvements involve reconstruction of tracks and/or bridges at 16 crossings in the Philadelphia area. The improvement avoids circuitous routing of 37 miles and cuts transit time by 5 hours. The reason for the cooperative working relationship is that the state of Pennsylvania and businesses that operate there are likely to benefit from double-stack improvement, since the more efficient container operations result in fewer trucks on area roads and bridges. Funding has been provided by CSX Railroad (about one-third), PennDOT, federal earmark funding, and FHWA improvement funds. The public-private partnership that provided public funding for the double-stack rail improvement has been crucial to the success of the project, which will be completed in 2013. See the DVRPC web site for more information: http://www.dvrpc.org. Example 27-2. Truck Turn Time Study The purpose of the turn time study, performed in 2010-11, was to analyze truck movements at the ports of Los Angeles and Long Beach to quantify the length and impact of trucking delays in picking up and delivering containers to the port. The project used GPS data from 250 dray trucks at the ports already equipped with GPS and capturing the data in the METRIS system as part of an earlier 2006- 08 study funded by U.S.DOT. A Truck Turn Time Stakeholders Group was formed and oversaw the project. It consisted of port operators and truckers and was key to keeping parties talking with each other. A consulting firm analyzed the data and made recommendations to the stakeholder group. Private organizations including Ability Tri-Modal Transportation Service Inc, Port of Long Beach, Port of Los Angeles, and PierPass, Inc, all contributed funding for the turn time study to pay for the consultant’s analysis and report. The analysis of 6 months of data from October 2010 to March 2011 showed where bottlenecks occurred and measured time that trucks waited in queues and time picking up and delivering containers in the ports. The analysis was documented in an April 2011 report by Digital Geo- graphic Research Corporation for PierPass, Inc. and Ability/Tri-Modal Transporta- tion Service. Efforts by stakeholders to implement the recommendations of the analyses continue. A METRIS web site has more information: www.metris.us.

Freight Data Sharing Guidelines 47 Guideline 28 – Consider Gathering Data from Volunteer Stakeholder Groups or Roundtables In addition to open source data usually available without cost, some projects obtain data through the network of stakeholder groups involved in the projects. Such data lowers cost and reduces the need to set up data sharing agreements. The Minnesota DOT, described earlier for its contribution to identifying publicly available data sources, included locally gathered data involv- ing project stakeholders as described in Example 28-1. Another example of the use of such data is the Greater Toronto-Hamilton Area Urban Freight Study, described in Example 28-2. Example 28-1. Minnesota FPM The Minnesota DOT study of freight performance measures involved data on all transportation modes. The project’s Technical Advisory Panel included members from both public and private transportation organizations. Operational data such as travel time data, loop detector data, and classification data were identified as good open source data, but it was found that these sources have not been tapped fully. The project was completed in July 2008 and showed that public-private approaches in gathering data, especially travel time along corridors, is a productive way to obtain data. The study also showed that such partnerships between public and private agencies and among different public agencies at different levels are criti- cal in developing understanding of freight data as freight flow is not confined to one jurisdiction. See http://www.lrrb.org/PDF/200812.pdf for more information. Example 28-2. GTHA Urban Freight Study The Metrolinx Greater Toronto and Hamilton Area (GTHA) Ontario Urban Freight Study, undertaken between December 2009 and January 2011, identified chal- lenges to strengthening urban freight in the region and summarized action to boost freight capacity and freight efficiency. A key aspect of the project was the involvement of stakeholders. Two stakeholder groups contributed. One was a Goods Movement Roundtable that included private sector industries and carriers, their associations, and marine port and airport authorities. The other was a Tech- nical Working Group that included representatives of local and regional govern- ments within the GTHA, as well as the provincial and federal governments, two Class I railroads, and two local universities. The project gathered information for the air, marine, rail, and highway modes, primarily publicly available data or infor- mation voluntarily presented by stakeholders. The freight data was collected in a series of meetings, workshops, and one-to-one interviews, including environmental scan information from stakeholders about freight conditions and issues. The proj- ect studied the current activities in urban freight and the conditions under which freight moves through the “last mile” portions of the journey. The summarized results were included in an overview report and a technical report, GTHA Urban Freight Study, FINAL DRAFT, Toronto, Ontario, February 2011 http://www.metrolinx. com/en/regionalplanning/goodsmovement/GTHA_UFS_Final_DRAFT.pdf.