Over the past decade, shared mobility services such as ride share, bike share, electric scooters, and car sharing have reshaped travel behavior in Minnesota, bringing both new convenience and network friction. Shared mobility improves urban accessibility and reduces individual vehicle ownership but also creates additional traffic with empty rebalancing trips (moving idle vehicles to where they’re needed) and congestion at curbs during pickups and drop-offs. Most transportation models can’t account for these dynamic interactions on traffic, greenhouse gas emissions, and regional access, creating challenges for transportation planners.
In a recent research project sponsored by the Minnesota Local Road Research Board, UMN Department of Civil, Environmental, and Geo- Engineering (CEGE) researchers aimed to address this challenge by creating a model that captures the full scope of shared mobility’s network impacts.
“For years, transportation planners have evaluated ride sharing, bike sharing, and transit in isolated silos, but on real streets these modes constantly interact,” says project lead Michael Levin, CEGE associate professor and CTS scholar. “Without a model that captures dynamic trade-offs, transportation planners were effectively flying blind when trying to manage curb space, anticipate traffic bottlenecks, or meet emission targets.”
To address this need, Levin’s team designed an integrated traffic system simulation of the Twin Cities metro-area road network. Rather than analyzing each mode in isolation, the new model uses more than 57,000 road segments, 24,000 intersections, and 11 daily demand periods to evaluate 6 travel modes — private vehicles, transit, ride hailing, car sharing, e-scooters, and bike sharing — simultaneously.
The model incorporates essential operational details, including how commuters weigh real-time trade-offs, fleet-size limits, walking distances to micromobility, and empty “deadhead” miles driven by ride-hailing vehicles between pickups. It also simulates how vehicles dynamically reroute under congestion and calculates changes in tailpipe emissions and greenhouse gases.
“By capturing these operational realities in a single framework, we can give agencies a clear, realistic picture of how these services actually impact local road networks,” Levin adds.
Next, the researchers used the model to assess the network trade-offs of shared mobility:
- Ride-hailing impacts: Ride-hailing services such as Uber and Lyft reduce parking demand and increase travel flexibility but elevate total vehicle miles traveled because of empty rebalancing trips. On certain corridors, these services also compete with public transit, drawing riders away from bus and rail routes.
- Car sharing and micromobility benefits: Shared electric vehicles (like Evie) and micromobility options successfully lower net greenhouse gas emissions and improve first- and last-mile transit access. However, system-wide environmental benefits depend heavily on the scale of the fleet and service-area boundaries.
- Accessibility enhancements: Shared options substantially improve regional travel time and access to critical destinations such as medical facilities, grocery stores, and employment centers — especially in areas with little transit coverage.
To evaluate how local transportation policies ripple through a complex transportation network, the researchers used the model to simulate a series of different policy scenarios. Their findings show how single-focused policies often result in unintended consequences. For example, imposing high fees on ride-hailing services cut local traffic delay by more than half but also increased carbon emissions, reduced accessibility, and decreased affordable mobility options — leading to longer trips in private vehicles. Conversely, incentivizing electric car sharing proved to be a dual-benefit strategy that reduced both congestion and emissions without sacrificing public access. Restricting or banning micromobility backfired, worsening traffic delays and emissions by severing critical first- and last-mile connections to public transit.
“The big-picture takeaway is that shared options work best when planned alongside existing public transit rather than left to operate on their own,” Levin says. “Our model shows how multimodal options can increase accessibility for residents without a private vehicle, but it requires teamwork between local government and private operators, like sharing trip data to manage curb space and locating scooter hubs right at transit stops.”
Elliott McFadden, MnDOT’s emerging mobility unit supervisor and the project’s technical liaison, noted that the modeling could help local practitioners “better understand the trade-offs of different levels of shared mobility services in their community.”
By pairing this new planning tool with targeted policies such as smart-curb management and electric fleet incentives, transportation practitioners could improve accessibility, mitigate congestion, and lower emissions without the cost and consequences of expanding road capacity.
— Megan Tsai, contributing writer
