As explored in previous posts in this series, transit’s foundational data standard is the General Transit Feed Specification (GTFS). This data standard has prompted many transit providers to open up their data. GTFS was created to describe fixed-route transportation services, though, and so “extensions” to GTFS have been added to address gaps in the original GTFS.
GTFS-flex is one of the most important extensions for mobility managers because it incorporates demand-responsive service and other, more flexible approaches to transportation services. For a thorough look at the history, status, and uses of GTFS-flex, see this white paper on GTFS-flex written by Weston Shippy and Thomas Craig of Trillium Transit for the National Center for Applied Transit Technology (N-CATT).
Work on this extension started in 2013, and by 2017, the Vermont Agency of Transportation (VTrans) launched the first trip planner that incorporated GTFS-flex data. The Federal Transit Authority provided funding for this project via the Mobility On Demand Sandbox program.
Like many if not all data standards, refinements to GTFS-flex continue. There is currently a draft proposal for GTFS-flex Version 2, which is likely to be finalized in 2022, if not sooner. The proposal and approval process is being managed by MobilityData, a nonprofit organization, which plays this important role for many GTFS extensions, as well as the General Bikeshare Feed Specification.
With a new version of GTFS-flex on the horizon, there is great promise for trip planning tools that will give a more complete picture of specialized transportation options. Accordingly, it’s a good time for mobility managers to assess their agencies’ readiness to open their data via a GTFS-flex feed. We spoke with Carol Schweiger of Schweiger Consulting LLC, who recently wrote a white paper on open source software and open data published by N-CATT, to get her perspective on the benefits of open data for mobility managers and steps to take before opening data.
Carol offered several benefits of open data:
- Awareness. “By putting open data out there, you can eventually increase awareness of your service.”
- Access. “Opening your data results in better access–for individuals, other agencies you already work with, and potential partners.”
- Coordination. Open data increases opportunities for coordination with other agencies and can make that coordination easier.
- Transparency. “Whether you’re a public or private entity,” open data “shows what agency spends money on, and shows how data is being used to meet transit goals.”
When it comes to your first forays into open data, Carol recommends approaching the task carefully. “It’s not an instant process. Be thoughtful, and take it slow. There’s work that needs to be done prior to opening data.” That work falls into two categories: data and organizational capacity.
On the data side:
- Look at your goals. Are you trying to coordinate your services with other providers? Create new partnerships? Something else? Whatever your goal is, keep it in mind as you look back at your data to identify gaps.
- Understand what data you have and what you’re currently doing with it. For example, whatever you have for scheduling, make sure you understand what’s in it.
- What’s the quality of your data?
As far as organizational capacity:
- Build your data literacy by looking at the open data others are producing.
- You need to be able to understand data–but that doesn’t mean that you have to become a programmer.
- If you don’t already have in-house capacity for data-related work, consider whether you will train, hire, or bring in a consultant who has data skills.
- Plan for time to monitor your data; you’ll need to fix errors and maintain how you’re putting data out into the world. Keeping up with data monitoring isn’t easy for a small organization. Monitoring is a hidden cost of open data, and even if you outsource this task, you have to have a liaison to that person.
- Collaboration with other agencies about your data also takes time and represents another hidden cost.
Visit the white paper for more information on the benefits and challenges of open data, examples and case studies, and associated costs. These topics are also covered for open source software.
In addition to Carol’s guidance, we offer one final caveat for any data-driven project: the more data you have available to you, the more tempting it is to measure everything. But a more targeted approach guided by specific goals and questions will help you get meaningful data—and will help you balance the workload required to maintain that data.