The rise of new data sources and determining how best to incorporate them into the transportation planning process is a major area of research and consideration. At the same time, the desire to integrate private and third party datasets into the regional and community planning process is associated with a number of challenges. The following list contains a number of specific challenges associated with evolving travel behavior (considering such influences as new modes and TNCs) and integrating new data resources into the planning process:  

  • Existing planning models do not account for new shared mobility technologies that have begun to significantly alter travel behavior.
  • The timescale of some regional planning activities is changing due to the rapid evolution in delivery of transportation services from private sector built and/or operated technology and other drivers.
  • Responding to environmental and social volatility is creating new needs for responsive evaluation frameworks and challenging established transportation planning methods.
  • Proliferation of geo-referenced data and networks is demanding better methods and conflation to reconcile inconsistent mapping.
  • Data sharing/accessing or private sector (proprietary) data – e.g., negotiating data sharing agreements. Local jurisdictions typically develop their own individual agreements and data sharing requests with shared mobility companies (opportunity lost to share with others).
  • Accessing and understanding data derived by interpretation and preprocess from third-parties. 
  • General lack of familiarity with the data that shared mobility companies collect and how these data might be used to inform planning, operations, and the management of the multimodal system.
  • Lack of detail in private sector data (e.g., O-Ds, geographic resolution, discontinuity in location data, missing information not sufficient to be integrated into regional modeling and forecasting).
  • Some required elements and classifications (e.g., trip purposes, accompanying travelers) are not explicitly recorded.
  • PII/privacy – e.g., where and when people travel can be revealed.
  • In the case of some forms of big data, the location information is generated by non-transport activity (e.g., during phone call, text message etc.). Therefore, it cannot be converted directly to mobility data for transport studies and often requires significant processing.
  • Comparing different datasets, even if they are related – e.g., different collection periods and spatial units (census data are usually available at the block level, while mobile phone data relate to individual cell towers). 
  • Absence of personal or socio-demographic information of the user, which are key inputs. 
  • Passively generated datasets lack ground truth to be validated against.
  • Addressing computer resource requirements.

In addition, interviews with a number of planning agencies revealed a number of common issues:

  • The provenance of delivered data products and the terms and provisions of third-party contracts can impact how data are used and the general usefulness of that data for specific purposes.
  • Linking information from multiple network referenced sources is still a major challenge despite development of such reference networks as ARNOLD. Functionality in these networks is still limited by structure.
  • Adoption of cloud-based computing for planning oriented groups is still a hurdle. This results from a range of concerns including past capital investment in in-house resources, concerns about data protection, uncertainty on costs and concerns about interfacing with cloud-based systems.  In addition, the “trust factor” of using cloud-based computing is still an issue for public agencies in general.  
  • Progress toward more centralized data storage and management is primarily limited by institutional barriers and technical development resources. 
  • The wide range of scripts, programs, and other analysis tools present challenges for agencies in knowing which tools work in certain situations and with certain data. Better documented tool use and operation, more formalized acceptance testing, and standardized access to a central library of proven tools could contribute to more effective use of these tools.
  • Agencies need guidance and support for integrating third party and commercial data resources. The operating scale of many of the new data suppliers points toward models such as national guidance and data clearinghouse activities to support their inclusion. With TNC suppliers, establishing broad agreements to provide certain standard data products at the national level may be necessary to ensure data quality and to support effective integration for state and local agencies.

Referencing Page: