This paper presents challenges faced by the transit industry in adopting new technologies and the use of associated data. 1 Transit ITS data is a valuable resource for transit agencies. There is great potential to use the data to create enhance effectiveness and efficiency and move towards more data-driven decision-making. A few key examples include use of CAD/AVL or APC data to:

  • Measure on-time performance and improve service reliability for customers
  • Provide valuable data on the distribution of actual running times by month, day, and time of day, which can be fed into the scheduling system to reduce recovery time, and thereby save operator hours
  • Monitor dwell time at intersections in order to select where Transit Signal Priority (TSP) would be of most benefit
  • Build ridership profiles by stop to identify key markets and transfer locations for route design, and where shelters should be located
  • Identify recurring patterns of missed runs, overloads, early departures, etc. 

But while transit ITS offers promising potential, the use of ITS data is often an afterthought. There are growing trends towards interfaces, coordination, and integration of ITS technologies between different transit agencies, including: regional real-time passenger information portals and displays, multi-agency advanced fare collection and connection protection, multi-agency advanced fare collection, transit signal priority, integrated traffic and transit control centers, and integrated corridor management. However, multi-agency coordination and / or integration of technologies will bring about new inter-organizational challenges. As urban transportation is being transformed through big data, mobility on demand, and connected and automated vehicles, much of the transit industry is poorly positioned to be a full partner in these transformations. Some of the biggest challenges are listed below:

  • Lack of rights to the data – ownership of data produced by technology systems does not always reside with the public transit system
  • Few agencies are able to provide broad access to the data within their organization, both due to technical limitations and usability concerns. Even if they could provide organization-wide data access, individual users would be challenged to make sense of the complex heterogenous data ​
  • Identifying and correcting data quality issues​
  • Organization and management of automated data within most agencies
    • Managing large amounts of data – Dynamic or highly complex data content​ requires some level of expertise to manage, troubleshoot, and visualize 
    • Lack of data retention policies
    • Lack of data inventory
    • Conflicts between data emerging from different systems (e.g. APC vs. AFC)
    • Missing or corrupted data
    • Creation of a data warehouse
    • Lack of diagnostic tools to determine cause of data collection/matching failures
  • Integrating different ITS sets together and with other sources​
    • Each system (e.g., AVL, APC and AFC from different vendors) has different database structures and data formats​
    • Wide variation in data types exist that are specific to individual applications, limiting the possibility for integration and extending the time needed for employees to gain a working knowledge of applications​
    • Multiple competing standards​ such that integration between different data systems or applications is difficult or impossible to achieve
  • Lack of internal expertise of data aggregation and visualization techniques or supplier-provided tools to match, clean, maintain, and analyze data to create information and standardized reports that provide meaningful information to support planning and operations decision making
  • Keeping pace with the data industry – even small changes to data structures, architecture, or standards requires time for agencies to comply with​
  • Misaligned objectives/motivations of the private sector (e.g., private sector processes and formats can differ from those used in the public sector, limiting their ability to integrate data or collaborate with transit agency partners)
  • Obtaining buy-in from decision makers on new approaches to sharing, collecting, storing, and managing data can be difficult and time-intensive. They do not always see the benefits of updating existing systems and processes​ – technology and data often take a back seat to ongoing operations and sufficient funding (the transit industry is more of an operations-driven culture than a data-driven decision-making culture).
  • Due to the lack of open data standards, guidance, and third party assistance, many agencies develop their own unique approaches and products, often “reinventing the wheel” unnecessarily​

With this comprehensive list of challenges comes a host of needs, which are summarized below: 

  • Better, more complete data for use in performance analysis and decision-making​
    • Understanding of how to integrate disparate datasets
    • Requests for proposals and subsequent contracts need to clearly state that all data produced by the technology systems will be owned by the transit system, not the supplier. In addition, they should provide for access to all raw level data, for purposes of diagnostics, or analysis.
  • Support and funding for open data initiatives to further spark adoption and innovation 
  • Promotion of open standards for real-time transit information with policies and incentives for vendors to adopt any open standard. This will promote more flexibility and shape future open standards, services, and systems development​
  • Guidance on how an agency can develop its own modern data practice, with processes and applications that help them assess data quality, manage dynamic and complex datasets, and visualize and make sense of the data (provide examples, best use practices, tools, guides, training materials, data dictionaries, validation tools)
  • Adequate resources and technical knowledge for pipeline development to process and analyze data and/or vendor development of open source data structures for smaller agencies lacking in-house expertise ​
  • The private sector needs clear incentives to collaborate with public sector organizations regarding data
  • Better methods of communicating to leadership the need for modern data management practices.

Collaboration between transit agencies is necessary to avoid duplication of effort in creating or recreating custom applications where functional solutions already exist. To be successful there needs to be a standardization or development process with clear business purpose, application, specificity, and versioning. To maintain momentum, define a governance structure that fosters a collaborative community for change development needs with discussion forums.

Referencing Page: