Conveyal
Published in

Conveyal

Access and Equity Impacts of Microtransit

A case study of new mobility-on-demand services and access to jobs in Los Angeles

Photo by Kenny Uong on Twitter

Many public transit agencies have launched on-demand services and “microtransit” partnerships in recent years. These services allow passengers to book rides in specific zones (e.g., via Transit App or a phone call). A driver then takes them to a fixed-route transit stop or destination in the zone, sometimes picking up or dropping off other passengers en-route. While “dial-a-ride” and paratransit services have run for decades, digital technologies have enabled a new generation of microtransit or mobility-on-demand (MOD) service to scale up. Proponents argue this service model is “nimble, flexible, and easily modified” and “seems to embody so many of the larger trends to emerge in transit, particularly since the COVID-19 crisis.”

Microtransit is expanding at agencies small and large. The Conveyal team has pioneered tools that help planners and the public understand its potential impact in terms of access to opportunities. For example, when the recent MVTA and SWT System-Wide Plan proposed expanding MVTA Connect and SW Prime microtransit services in Minnesota, Conveyal published a comparative journey mapper that integrated these services. Similarly, the WeGo Better Bus plan in Nashville introduced multiple MOD zones as part of a holistic, equity-focused redesign of their bus network. For more details on how WeGo planners used Conveyal features for origin-destination travel time equity analysis and public engagement focused on access metrics, see the TCRP Synthesis “Assessing Equity and Identifying Impacts Associated with Bus Network Redesigns.” On the West Coast, recent MOD pilots in the Los Angeles and Puget Sound regions have been studied well.

The evolution of microtransit in Los Angeles County provides a valuable case study for evaluating access and equity impacts. Below we recap the timeline of these changes and demonstrate how to analyze them with Conveyal.

Brief video advertising the passenger experience on Metro Micro

Evolution of Metro on-demand services

The Los Angeles County Metropolitan Transportation Authority (Metro) has been incrementally expanding its MOD and microtransit services for more than three years:

2019

  • January: MOD pilot is inaugurated, providing first-/last-mile connections with rail stations in three zones (Artesia, El Monte, North Hollywood)
  • May: Boundaries for three initial zones are expanded

2020

  • March: Due to COVID-19, first-/last-mile requirements for MOD service are dropped, allowing passengers to make direct point-to-point trips within zones. Metro also shifted available resources to provide food delivery to vulnerable families, in partnership with Best Start.
  • December: New microtransit service called Metro Micro is launched (in partnership with RideCo) in two zones (Watts/Willowbrook and LAX/Inglewood) in conjunction with major bus service changes (NextGen Phase 1)

2021

  • January: MOD zones are converted to Metro Micro service, yielding five zones total
  • June: Service is expanded to two additional zones (Pasadena and Highland Park). Soon after, Metro notes growing popularity and waiting times exceeding the 15-minute goal.
  • September: Two previous zones are combined, and an additional zone (North San Fernando Valley) is added in conjunction with major bus service changes (NextGen Phase 2)
  • December: Eighth Metro Micro zone (Westwood) is added
Animation showing MOD and Microtransit zones added over the years, plus an example extension scenario, loaded as Conveyal Data Sources

Measuring increased access to jobs

A previous article published in APTA Passenger Transport examines Metro’s original 2019 MOD pilot. This article included an access-to-jobs analysis powered by Conveyal’s open-source software, and it ended by suggesting an evaluation of the point-to-point service introduced in 2020. The results below build on that article, using recently released Conveyal features and Metro service snapshots from 2020 and 2021.

There are numerous reasons to use access to opportunity metrics in transportation equity efforts (as we argue in our previous blog post). The Conveyal user manual now includes guidance on summarizing results for entire regions and specific demographic groups. One way to promote equity is ensuring that enhanced service prioritizes both areas and people who have historically borne such burdens.

Specifying on-demand scenarios in Conveyal

The built-in scenario editor in Conveyal makes it easy to specify on-demand services. For example, after specifying microtransit zones as data sources, you can create a “Pickup Delay” modification and enable various experimental settings (as explained in the latest Conveyal user manual). With these flexible configuration settings, you can choose whether point-to-point service within specified zones or only first-/last-mile service to/from specific fixed-route stops is allowed and how overlapping zones (e.g., the overlapping Artesia and Watts/Willowbrook zones from 2020) should be handled.

Once your microtransit modification is ready, you can combine it with other network modifications in the Conveyal scenario table.

Results

With a scenario specified, you can run a regional analysis to evaluate its impacts in terms of access to opportunities. Conveyal’s automatic cloud scaling computes indicators for millions of origins in just a few minutes, including multiple opportunity types and travel time cutoffs. The results below show the number of jobs reachable within 60 minutes, from origins in a central part of Los Angeles County, using the 25th percentile of door-to-door travel times (including walking, waiting, and public transit) for journeys starting between 7:30 am and 8:30 am.

Conveyal lets you compare regional analysis results, which facilitates evaluating scenarios against a baseline. For example, the map below shows the increase in job access with point-to-point service enabled in the three original MOD zones, relative to 2019 service before the MOD pilot. Note that the largest increases (areas in darkest blue) roughly correspond to the boundaries of the three original zones (North Hollywood, El Monte, and Artesia/Compton). Within those zones, access gains are lower near existing rapid transit stops (where baseline access is already high). And access gains extend well beyond the zone boundaries, as people originating in other parts of the city can ride fixed-route transit, then take MOD service as a last-mile connection to jobs within the zones.

Change in job access with point-to-point service enabled in the three original MOD zones, relative to 2019 service before the MOD pilot

Conveyal’s flexible data source options let you upload your own shapefile layers and then convert them to aggregation areas that define regional summary and demographic indicators focused on equity. For example, you could summarize results for Metro’s Equity Focus Communities or the US CEQ’s Climate and Economic Justice Screening Tool zones. We use the California Healthy Places Index (HPI) in this case study. In the top HPI quartile areas, residents face relatively few environmental justice issues; conversely, in areas classified as HPI quartile 4, residents may bear disproportionate burdens of past transportation decision-making.

Toggling between the top and bottom HPI quartiles as aggregation areas in the Conveyal interface, you can observe that for the comparison shown above, point-to-point service in 3 MOD zones increased average job access by 2.8% for workers living in HPI quartile 1 and by 3.8% for workers living in HPI quartile 4. The larger increase for the lowest HPI quartile arguably indicates that service in these zones aligns with equity objectives.

Distribution of job access for workers living in the highest and lowest Healthy Places Index quartile areas

Similarly, we compared services in the following two years to the pre-MOD baseline for workers overall and workers living in areas in the lowest HPI quartile in this central part of Los Angeles County.

The December 2020 fixed-route service cuts led to areas with decreased access to jobs, but the average worker saw an 11.0% increase in job access, and the average worker living in the lowest HPI quartile saw an 11.4% increase in job access.

Change in job access, December 2020 relative to 2019 service before the MOD pilot

In December 2021, access gains were more widespread due to substantial restoration of scheduled services. The average job access increase, relative to 2019, was 19.2% for workers overall and 17.4% for workers living in HPI quartile 4 areas. While MOD and microtransit zones introduced in previous years clearly furthered equity goals, as indicated by the higher access gains for the bottom HPI quartile, the zones added in 2021 may not align as strongly. On other metrics, however, such as actual ridership, there may still be a strong equity case for the more recent zones; this topic would be worth further analysis. Given the spike of missed trips due to workforce unavailability and COVID-19 exposures in late 2021, it would also be worth considering reliability adjustments for this period (see our past post on this topic).

Conclusion

This analysis considered only a limited type of benefit — increased access to jobs. Access to healthcare and other opportunities, at other times of day, could be considered with the same analysis workflow. Additionally, a complete equity analysis would also need to consider other benefits and the costs, financial and otherwise, of microtransit services. With smaller vehicles and demand-responsive dispatching, microtransit cost and service levels are especially sensitive to demand spikes. While our analyses show potential increases in job access, realizing these increases for large numbers of passengers with lower capacity on-demand vehicles might be expensive. If you are interested in pairing Conveyal’s detailed access to opportunities capabilities with other analyses of microtransit performance, please get in touch.

For future research, it would also be interesting to test a range of extensions to the MOD zones. As a final example, we extended the Pasadena and Highland Park zones to cover one adjacent light rail station each (Fillmore and Lincoln/Cypress). As shown in the map below, this example future scenario leads to clusters of increased job access; some clusters are located in surprising places. Clusters in the Pasadena zone can access more than 100,000 additional jobs within one hour if microtransit lets them connect to light rail at Fillmore, one station closer to Downtown. Origins in Boyle Heights, which are not within a microtransit zone themselves, also see notable job access gains, thanks to more convenient last-mile microtransit options from the added Lincoln/Cypress Station.

Change in access to jobs, example extension alternative (with microtransit service added at the two stops shown) vs. December 2021. Note that access increases for various origin clusters, including some outside the microtransit service zones.
Evolution of access to jobs, including an example future scenario.

The complex patterns in these results underscore the value of a detailed, rapid-turnaround multi-modal routing platform like Conveyal. Network effects often lead to access gains in unexpected places, and rapid-turnaround capabilities for assessing multiple scenarios help planners iterate through alternatives and design services that advance their access and equity goals. The Conveyal team is excited to continue developing these capabilities, and we look forward to hearing from you about new ways to use them.

--

--

--

We enable people and institutions to make smarter decisions by applying technical expertise in open source software and open data to create improved transportation systems, more livable cities, and a better world.

Recommended from Medium

You know your favorite music style, but what about your conference style?

A job ready Data Science curriculum

PhD student in Data science who interested in Stock price prediction topic

Learning from Eisenhower, Aristotle and Toyota

Can we use SQL for advanced analytical techniques?

Deduplicating records — is Machine Learning the answer?

Duplicate customer records

Spot Content Ideas Using 4 Free APIs In Python

Build Your First Data Quality Checking Web App in Python Using Streamlit

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Anson Stewart

Anson Stewart

Analysis and Research, @conveyal | PhD in Transportation, @MIT | '10 TJ Watson Fellow + @SwatAlum | Californian in exile on East Coast

More from Medium

AI-Driven Photo Editing Platform of Batch.ai

Impersonating a Google cloud plattform Service account

[Spotlight] Promoting collaboration at the national and local level to strengthen data governance

DataGovs now supports multi-cloud data storage for better data governance