MXD Trip Generation Model

A Tool for Estimating Travel Demand from Mixed-Use Developments

Traditional trip generation methods often overestimate traffic from mixed-use development. Standard Institute of Transportation Engineers (ITE) trip generation rates treat each land use independently and assume most trips arrive by car (when applied without adjustment for mixed-use context). In reality, mixed-use places function differently: people walk between uses, combine trips, or arrive by transit or bike.

Aerial video showing cars and people moving through a long corridor with residential, urban and rural areas.
Pedestrians walking in front of a mixed-use building - showing both shopping on the lower levels and residential units on the top levels.

Built on Research and Practice

2010 EPA Partnership

Over a decade ago, we first partnered with the US Environmental Protection Agency to develop the original MXD tool.

Continued Advancement

We have continued refining the model through research, project applications, and validation efforts.

Latest Update

The newest version incorporates trip generation rates from the ITE Trip Generation Manual, 12th Edition (2025).

More Accurate Trip Generation

MXD builds on conventional ITE trip generation methods to estimate trips for mixed-use development, accounting for internal trips between complementary land uses and how the surrounding environment influences travel by walking, biking, transit, and driving. By incorporating built environment variables, often referred to in transportation planning as the “D’s” (such as density, diversity of land uses, and design/connectivity), MXD captures how land use mix and site context shape travel behavior.

The enhanced model evaluates seven key variables that influence trip-making patterns, allowing it to more accurately represent how different types and scales of development generate travel. The result is a more realistic picture of how people move within mixed-use environments, helping address a common bias against mixed-use development in traditional traffic analysis and producing more accurate traffic estimates.

Real-World Data, Validated.

We validated the MXD model using mixed-use sites across the US.
During the critical weekday PM peak hour, MXD predictions averaged approximately 3% deviation from observed traffic counts, compared with about 10% deviation using the traditional ITE-based methodology (NCHRP 684). These results demonstrate the value of incorporating built-environment context into trip generation analysis.
Read The Paper
Graph showing that MXD more closely estimates trip generation when compared to other models.

Customizable for Communities

Because travel behavior varies by region, we can develop locally calibrated versions of the MXD tool tailored to a community’s development patterns and transportation goals.

This allows agencies to evaluate mixed-use development using methods that better reflect local travel behavior and planning goals.

Customizable for Communities

Because travel behavior varies by region, we can develop locally calibrated versions of the MXD tool tailored to a community’s development patterns and transportation goals.

This allows agencies to evaluate mixed-use development using methods that better reflect local travel behavior and planning goals.

Man on a bike riding down a bike lane while a woman walks on the sidewalk in the city.

Applied in Practice

Washington, DC

We partnered with the District Department of Transportation (DDOT) to develop a customized MXD tool tailored to local conditions. The collaboration resulted in an enhanced tool that helps DDOT evaluate multimodal travel behavior during development review.
Project deliverables included:

Customized MXD model for DDOT

User guide for agency staff

Supporting literature review

What’s Next

In 2025, we conducted one of the largest mixed-use traffic data collection efforts to date, studying 37 mixed-use sites across the US. Findings from this dataset will be released in 2026 and will provide new insights into post-COVID travel behavior, further refining MXD calibration.

Interested in applying MXD for your community or project? We can develop locally calibrated tools that support more accurate and context-sensitive transportation analysis.

Let’s Connect

Headshot of staff member John Gard

John Gard

Principal

Email Me