TransitLink Delivery

Local restaurants pay a huge platform fee to Doordash which makes it hard to survive for smaller businesses. We propose using the extremely well connected network of ASUCD Unitrans buses for delivery

Comment

Inspiration

Downtown Davis is powered by independent restaurants and retailers, yet many are squeezed by high third‑party delivery fees. The DDBA’s mission is to sustain the economic vitality of 32+ blocks of local commerce through advocacy and community support. At the same time, the ASUCD Unitrans network is extremely well connected, running through downtown and reaching most of the city. A local delivery alternative is essential to keep revenue in the community, strengthen the downtown business ecosystem, and leverage that existing transit backbone so small operators can compete without sacrificing their margins.

What it does

TransitLink Delivery is a DDBA‑hosted, restaurant‑friendly delivery platform that combines direct ordering pages with a coordinated local delivery network. We leverage the extremely well‑connected ASUCD Unitrans bus network that runs through downtown and reaches most of the city, turning existing transit capacity into a sustainable alternative. Customers place orders and share their location; the system assigns the nearest downtown‑serving Unitrans stop and a compatible route, enabling reliable pickup and delivery with live status updates and QR‑verified handoffs. Live Unitrans vehicle data is available via an open public feed, which we use for real‑time order tracking. Restaurants keep more of their revenue while students and residents get predictable, affordable delivery that supports local businesses.

How we built it

We built a full‑stack system with a web app for customers, restaurant owners, and stewards, plus a backend that manages orders, routes, and delivery status. The platform integrates Unitrans data to assign routes that serve downtown, uses a local stop‑based handoff model, and provides live status tracking once orders are on the bus. Role‑based workflows ensure restaurants accept and prepare orders, stewards load and complete deliveries, and customers track progress with verified pickup.

Challenges we ran into

  • Translating a downtown‑focused delivery requirement into a practical route and stop assignment flow.
  • Balancing accuracy in live tracking with limited real‑time data availability.
  • Coordinating multiple roles (restaurant, steward, customer) with clear, secure status transitions.
  • Designing a location‑based pickup experience that remains simple and reliable.

Accomplishments that we're proud of

  • A DDBA‑aligned delivery model that keeps restaurants in control and reduces commission pressure.
  • A location‑based flow that automatically selects a downtown‑serving stop and route.
  • Live, role‑aware order tracking with QR‑verified handoff for accountability.
  • A clean multi‑role experience that fits the day‑to‑day reality of local restaurants.

What we learned

We learned how important it is to align technical design with community constraints. Downtown‑first delivery isn’t just a routing problem; it’s a business sustainability problem. Building around a local transit network creates a viable alternative to national delivery apps, but only if the experience is simple for restaurants and transparent for customers.

What's next for TransitLink Delivery

  • Expand the downtown route map and validate it with DDBA and Unitrans stakeholders.
  • Add operational analytics for restaurants and DDBA to measure savings and impact.
  • Pilot the workflow with a small set of partner restaurants and iterate on handoff timing.
  • Explore incentives for stewards and local delivery partners to ensure coverage during peak hours.

Built With

+ 1 more
Share this project:

Updates

posted an update —

Hi, we’re Team TransitLink. We convert existing campus buses into predictable, low-cost delivery vehicles by integrating directly with live transit routes.

The key difference is that we’re route-aware, not distance-based. Buses don’t move freely — they follow ordered stops. So instead of using straight-line distance, we ingest live GPS and route XML, map each bus to its nearest stop using a spatial index, and compute ETA based on the number of stops remaining. Since campus routes are circular, we handle forward and backward traversal safely to avoid rollover errors.

ETA is calculated as stops_between × average_minutes_per_stop, and if route data fails, we gracefully fall back to distance-based estimation with a lower confidence score.

For matching, we evaluate every active bus in O(B) time and score them based on distance to restaurant, delivery ETA, alignment with prep time, and route efficiency. The highest score wins. Matching is stateless, so it scales horizontally.

Operationally, it’s simple. Orders are prepaid, so restaurants have zero risk. The steward scans a QR at pickup — status becomes ON_BUS. They scan again at delivery — status becomes COMPLETED. Two scans, two timestamps. That’s the full audit trail.

Authentication is role-based. Students see only their orders, restaurants see their queue, and stewards see assigned deliveries. No cross-access.

We also ingest live bus tracking roughly every 15 seconds, so ETAs update dynamically — but because they’re anchored to stop indices instead of raw GPS, they’re stable and not jittery.

TransitLink works because we respect routes, ensure accountability through QR verification, and reuse existing infrastructure — making delivery predictable, frictionless, and scalable

Log in or sign up for Devpost to join the conversation.