Rethinking Ridership: Making Everyday Modeling and Optimization Possible

For decades, travel demand modeling has been the ultimate “what if” tool for transit planning — powerful in theory, but rarely practical in everyday use. It’s how agencies can see into the future, anticipating how riders will respond to service changes before they’re made.

But the reality is that most planners have been shut out of that process. Traditional modeling systems are expensive, slow, and dependent on specialist expertise. Running a single scenario can take months. For many agencies, it’s a tool used once a decade — not once a week.

Even worse, the data that powers these systems often lags behind the real world. Many agencies are still working from household travel surveys that predate COVID, trying to plan for travel behaviors that no longer exist.

It’s time to bring modeling — and optimization — into the hands of the people who need it most: planners making decisions every day.

From Static Models to Living Systems

At Ontra, we believe that travel demand modeling should be dynamic, accessible, and actionable — not a black box or a one-off consultant report.

That’s why we built Ridership Predictions, a next-generation modeling engine integrated directly into the Ontra platform. It gives planners the ability to forecast how people move, test changes in real time, and even optimize their networks — all in one place.

With Ridership Predictions, planners can finally ask and answer the questions that define modern transit networks:

• How will ridership respond if we shift service from low-performing routes to high-demand corridors?
• What’s the tradeoff between wider coverage and shorter waits?
• How do small changes — an added trip, a relocated stop, a service-hour tweak — ripple across the network?

The result is a workflow where modeling becomes part of planning itself, not a separate, siloed exercise.

How It Works

Ridership Predictions combines three components that together mirror how people actually travel:

Demand Model — Where people travel

Fusing your agency’s ridership and APC data with census and modern, anonymized mobility data, Ontra builds a real-time map of where trips begin and end — no more relying on outdated surveys.

Routing Engine — How they move

Using your GTFS and any proposed scenarios, Ontra simulates every possible journey — across walking, waiting, transferring, and riding — to understand what options riders truly have.

Behavior Model — What they choose

A machine learning model predicts which journeys people are most likely to take, balancing travel time, reliability, walk distance, and the number of transfers.

Together, these models generate a complete forecast of ridership, accessibility, and travel times across your proposed network, all compared directly to your baseline.

Bus Route Analyzer

Our bus route analyzer pairs Trip Flows and Load Factor to help you dive into current route performance and understand how we expect it to change.

Trip Flows quantifies the number of trips between selected stop clusters, revealing origin–destination patterns and demand hotspots.

Load Factor overlays a color‑coded gradient along each route segment to show crowding and spare capacity on the map.

Together, these tools help you quantify the impact of proposed changes on ridership, usage patterns, and core service performance metrics such as load factors, reliability, and travel times.

Example: Finding the Right Balance in Riverton

Consider Riverton, a growing mid-sized city. Downtown bus corridors are at capacity, while outer neighborhoods are underserved. The classic debate emerges: Should the agency double down on frequency, or expand coverage?

With Ontra, the planning team can test both ideas — quickly and transparently.

The Frequency-First Plan doubles service on the busiest corridors, at the cost of trimming a few low-performing routes.
The Coverage Plan adds a new route to reach an underserved job center.

Running Ridership Predictions takes minutes, not months. The results are clear: frequency investments drive a 5% ridership gain systemwide, while the coverage expansion improves access but not total ridership.

Combining elements of both produces the best overall outcome — a 3% ridership increase with improved access and reduced average travel time.

What was once an abstract tradeoff now becomes a quantifiable choice.

Or, Let Ontra Optimize It for You

In many cases, planners don’t just want to compare scenarios — they want to know which combination delivers the best results within their real-world limits.

That’s where the Ontra Optimizer comes in.

Rather than manually testing multiple versions, planners can define their resources and goals — like total budget, available vehicle hours, or target ridership — and let Ontra automatically search for the optimal configuration of routes, frequencies, and zones.

Under the hood, Ontra’s optimization engine evaluates thousands of possible network permutations to identify those that maximize performance under your constraints.

You can set objectives such as:

• Maximize ridership while staying within your operating budget
• Minimize average travel time subject to equity targets
• Balance frequency and coverage for maximum access

The result is a data-driven service plan that meets your strategic goals — without guesswork or endless manual iterations.

Modeling + Optimization: A New Planning Standard

This is more than just faster modeling. It’s a fundamentally different approach to planning — one where data, simulation, and optimization are part of the same process.

With Ontra, planners can move from reacting to problems to designing better networks continuously. Modeling becomes not just a technical task, but an everyday planning habit — accessible to every agency, no matter its size or staff capacity.

When planners have tools that can both predict and optimize, decisions become clearer, tradeoffs become transparent, and service becomes smarter.

That’s the vision behind Ontra Ridership Predictions — and it’s how we’re bringing the next generation of transit planning to life.