Create and export GTFS-Flex Data with Spare's free toolLearn More

Go Home
Back to Stories

Customer story

Using transit simulations to optimize a subsidized taxi service


To better understand the potential of on-demand microtransit in Collin County and identify suitable operational models.


The McKinney Urbanized Area, which is part of the Dallas-Fort Worth-Arlington metropolitan area, offers its residents a voucher-based, subsidized on-demand taxi service through Collin County Transit (CCT). The service covers a rather large area and includes the cities of Celina, Lowry Crossing, McKinney, Melissa, Princeton and Prosper. Because it is an unpooled, single-operator service, there are opportunities to provide better service in a more efficient way.

CCT was interested in the potential cost savings of moving to a fully-automated on-demand microtransit platform delivered through dedicated vehicles or a third-party. It also wanted to find a way to seamlessly connect its service to the commuter transit offered by Dallas Area Rapid Transit (DART).

In order to best understand which options would be most advantageous, CCT asked Spare to simulate different scenarios using its advanced simulation tool, Spare Realize. The learnings from that process have helped CCT to decide what resources it should dedicate to microtransit, what type of service would help it meet its goals and the potential trade-offs of certain choices.

Collin County Transit
Collin County, TX
July 2020
Service type


  • With only one operator and a handful of vehicles dedicated to the on-demand taxi service, the program suffered from supply issues;
  • In the single-operator model, the provider was paid on a per trip basis leading to relatively high per trip costs;
  • CCT wanted to explore the implications of moving to a flat fare structure compared to its established variable fare;
  • The on-demand taxi service is not integrated with different modes making it difficult for residents to connect to DART light rail and the broader fixed-route transportation network.


Spare leveraged demand data from its nearby service with DCTA and DART, as well as CCT’s own data, to run 54 different simulations around on-demand microtransit in the CCT coverage zone using its tool Spare Realize.

The simulations covered everything from where the microtransit system should operate, whom it is likely to serve, when it should operate, the service model it should follow (stop-to-stop, door-to-door), and its potential costs and returns. Based on the different scenarios and data processed by the Spare Realize algorithms, Spare was able to make specific recommendations on operational efficiency, cost, service optimization and agency goals, which CCT stakeholders could easily understand and evaluate.

"Thanks to Spare, we were really able to understand not only the impact of moving to on-demand microtransit but also all the different ways we could configure the service to meet our goals. That kind of insight is invaluable to transit planners and often the missing piece for getting stakeholder buy-in. The simulations with Spare weren’t just an exercise; they were a vital step in ensuring the future success of transit in our county.”

Akia Pichon
Akia PichonTransit Administrator, CCT


Using simulations, Spare concluded that there was high potential for excellent microtransit in CCT’s coverage area based on two models: a more expensive, door-to-door service using only dedicated vehicles, and lower cost option that relied on more pooling, a mixed supply of vehicles and a stop-to-stop configuration. Though the second option would be disadvantageous for low-mobility rider groups, Spare demonstrated that by segmenting riders in its software platform it could take a two-pronged approach.

Spare suggested that CCT extend coverage until midnight to accommodate those requiring late night transportation, which would spread demand more evenly throughout the day, increase ridership and advance CCT’s mission of providing more equitable transit across the county. This would boost ridership by 15 percent. It also demonstrated how an automated on-demand microtransit service would encourage riders to connect to DART by essentially turning it into one integrated, seamless multi-modal system.

Estimated pooling rate
Estimated cost savings per trip
Recommended operational model