Over the past two years, the Sacramento Area Council of Governments (SACOG) has been pilot testing a powerful new data platform in a partnership with Caltrans and the California Air Resources Board. They call it the Big Data for Transportation Planning pilot project. What it lacks in a catchy name, its gains in providing a new –and more current–way to analyze regional travel trends.
SACOG and the state agencies jointly procured Replica, a data platform for the built environment. (Disclosure: Replica also sponsored CALCOG’s 2021 Regional Leadership Forum.). Replica uses machine learning to turn many streams of raw de-identified data inputs —(e.g. anonymized location data from smartphone apps, and cellular devices, credit card point of purchase data, and others)— into valuable insights useful for transportation modelers, planners, and policy-makers.
Trends & Places
Replica services are built around specific platforms.
The “Trends” platform can provide a weekly snapshot of activities across the region. These trends can also be viewed at the state, county, and even census tract level. A snapshot includes the numbers of daily trips, vehicle miles traveled (VMT), trip start times, and trips to work and school.
The “Places” platform allows planners to do road link and transit line analysis, showing information such as trip origin, destination, length of trip, trip mode, breakdown of travelers, household income of transit line users, and even the reason for the trip. Replica also provides spending data across sectors such as gas stations, grocery stores, and restaurants and bars. Race and ethnicity data are also included in the available classifications of travelers within Replica, though application of these data still need more testing and evaluation.
How Replica Protects Anonymity
The partnering agencies placed a high priority on privacy and confidentiality. It is important to note that Replica’s outputs do not release raw data. Instead, Replica turns the raw data into synthesized “replicas.” That way, no actual raw geolocation data of individuals is accessible, making tracking individual movements impossible. Only the synthesized outputs are shared.
More specifically, the Replica platform uses American Community Survey (ACS) data and other sources about who lives in a given area as well as regional housing and employment availability. See detailed listing of data sources. It then applies modeling and optimization algorithms to generate a representative population that is statistically equivalent to the census population, and then matches the population to “personas,”, which are an amalgam of synthetic travel and activity patterns based on raw device data. These representative people and households are assigned housing units and locations of workplaces and schools. Personas extract behavioral patterns from de-identified mobile location data collected from mobile devices of real people. In this way, the source data is aggregated and de-personalized so the resulting insights are anonymized.
Back to Planning
The first step for planners using Replica is to validate the tool’s synthetic data by comparing it to on-the-ground observed data. SACOG’s modeling team validated it against Caltrans’ traffic data, population data, and data from local agencies and transit operators, and found it to be good for a variety of applications. Replica also tags its own insights with measures of how confident it is about the data.
The ground-truth validation process described above was a minimum threshold for SACOG and its state partners. SACOG has used it to gain insights about their projects. The Places feature provide , which shows information in quarters from fall 2018 to fall 2020. For example, using filters and layers, planners can do road link analysis and transit line analysis, showing information such as trip origin, destination length of trip, trip mode, ethnic breakdown of travelers, household income of transit line users, and even the reason for the trip.
This kind of data is relatively new, so it is difficult to go back farther than a couple of years to compare conditions because the data did not exist “way back then.”
Analyzing VMT: California’s SB 743
Replica’s tools can be calibrated to help assess the transportation aspects of new projects under SB 743. Instead of traffic congestion, planners must now analyze the effect a project will have on the amount of vehicle miles traveled (VMT). SACOG engaged Replica to make such comparisons. As a result, Replica produced region-wide person travel patterns for multiple seasons. Users can then use the generated trip tables to help establish existing conditions before investing in large-scale projects.
The data allows SACOG to better understand the VMT patterns of its residents and of its road network, and to calculate VMT for trips that occur outside of its region. It also allows them to understand the breakdown of VMTs across different trip purposes like commuting, school, shopping, and other destinations.
Pandemic Travel: Week by Week
Even though the pilot was launched before the emergence of COVID-19, the model was able to show the disruptive effect of the shelter-in-place rules. Working from home. Distance learning. Takeout or delivery instead of going out to eat. Home delivery instead of shopping. And on, and on. Replica captured the main changes in travel patterns observed during the pandemic (see the graph below).
Replica also provides estimates of transit trips by trip origin, destination, and traveler’s demographics. This feature helps transit planners identify popular transit lines, origin, destination, locate potential transit users, and prioritize transit funding. The graph below shows the most popular transit destinations which were identified in SACOG Next Generation Transit pilot, based on Replica data.
It’s clear that big data will play an increasingly larger role in planning and development decisions. “Big Data holds the promise to help transportation planners to better understand how travelers are adapting to the changing world we live in,” says SACOG’s Data and Analysis Manager Bruce Griesenbeck. He adds that Replica was allowing his team to learn where Big Data was useful — and equally important, where it may not be.
Replica agrees. “Throughout our partnership with SACOG, we’ve learned so much and are eager to partner with other agencies to help gain insights about their own projects,” said Replica CEO Nick Bowden.
For More Information
- SACOG: Big Data Pilot Project
- SACOG: Replica Pilot Explainer Deck
- Replica FAQs (posted on SACOG website): Good explainer of Replica data and platforms
- Big Data: A Conversation with Replica CEO Nick Bowden #7; CALCOG’s Regional Leadership Forum (2021)
- Replica Article (on Medium): Reveal Equitable Change in your City
- Replica Article (on Medium): Yes, People may be Moving Again . . . But are They Moving the Same Way?
- Replica Article (on Medium): Building a Bike-Forward Future