A summary of the most common questions that our customers ask about our data.
Data Collection & Accuracy
Where does Urban SDK obtain traffic data?
A: Traffic speed and delay data are obtained through vehicle telemetry data with a nationwide penetration rate of 20-35% of vehicles, while traffic volume data such as annual average daily traffic (AADT) and annual average hourly traffic (AAHT) are modelled using a mixture of telemetry data and state department ground truth data. The model estimates volumes on all public roadways applying industry accepted mean average error (MAE) standards to ensure quality output. Mean average error differs state-by-state based on the availability and quality of ground truth data, contact your sales rep if you would like information on accuracy in your state.
Where does Urban SDK obtain collision data?
A: Urban SDK sources data from state-specific databases, such as Florida’s SunGuide and Tennessee’s TDOT for collision data but are able to integrate with any state database should access be provided. For fatality information, the federal FARS database is relied upon, with a typical one-year delay. Collision data is integrated as points into Urban SDK using each collision record's latitude and longitude.
What should I do to validate the accuracy of Urban SDK's data in my community?
A: Urban SDK offers data validation services, including methodology white papers, government validation reports, and tailored validations. To conduct a data validation, clients need to provide recent study details, including date, time, and ground-truth counts for comparison.
Why might Urban SDK's speed values differ from my ground truth values?
A: Urban SDK traffic data and tube counts differ in methods. Tube counts offer short-term snapshots with high sample rates, while Urban SDK collects long-term data at a 20-35% penetration rate. Tube counts are more accurate for short-term trends over a day or two, whereas Urban SDK data is better for medium-to-long-term trends with a larger sample pool over time.
Traffic Data Questions
Can average annual daily traffic (AADT) be provided by day and period or hour of day?
A: Yes, we offer Annual Average Hourly Traffic (AAHT) volumes which is an annual value broken down by typical day of week, period of day, and hour of day. In Aggregated Insights reports where data is provided by period of day, the AAHT value represents the max hourly volume within the selected period(s). For Daily Insights reports where data is provided by hour of day, the AAHT is provided as the calculated annual hourly value.
How does Urban SDK estimate day of week and hourly volume distributions?
A: Urban SDK estimates these distributions by using volume distribution percentages from ground truth sources, combined with vehicle telemetry data. These percentages are then applied to the calculated Annual Average Daily Traffic (AADT) value for road links, enabling Annual Average Hourly Volume (AAHT) breakdowns to be calculated by day of week, period of day, and hour of day.
What year is the traffic volume (AADT/AAHT) data based on?
A: AADT/AAHT data update timing is impacted by when state counts are released. Urban SDK's model typically reflects the previous year’s counts (I.e. 2022 AADT is available in 2023).
Can Urban SDK provide turning movement counts (TMCs)?
A: Urban SDK does not currently provide TMCs due to data accuracy limitations with current technology. Urban SDK is conducting research & development to determine the feasibility of quantifying the distribution of trips by intersection approach (i.e. 20% left, 50% through, 30% right) using traffic data, in combination with road geometry characteristics such as approach lane layout, intersection control, and pedestrian crosswalks as features to assist in model estimation. Research & Development is expected to conclude sometime in 2025.
What time period does Vehicle Miles Traveled (VMT) represent?
A: Vehicle Miles Traveled (VMT) is calculated annually using AADT (Annual Average Daily Traffic). Since AADT is an average for the year, VMT is also an annual measure and doesn't show monthly changes. The calculation multiplies the length of the road by the number of vehicles and by 365 days to estimate total miles traveled on the road annually.
What is Urban SDK Speed Limit?
A: Urban SDK Speed Limit is a model which provides a best estimate of posted speed limits on road links for which no ground truth speed limit is available. We use a combination of the road functional class, speed limit information provided by our raw data provider, and off-peak observed average travel speeds to estimate Urban SDK Speed Limit. As we continue to receive ground truth speed limit data from our customers within their boundaries, we are refining and updating Urban SDK Speed Limit to improve our level of accuracy in estimating posted speed limits with this metric.
Where do you get your speed limit data from?
A: Urban SDK provides an estimated Speed Limit. Customers also have the capability to integrate their own speed limits to replace Urban SDK's estimated speed limit. Speed Limits can be integrated within the Organization's Location Management settings.
How do you calculate Speed Limit Difference layers?
A: For customers with integrated speed limit data, Speed Difference layers will use those integrated speed limits. For customers without integrated speed limits, the layers will use Urban SDK's Speed Category. If some streets don't have updated speed limits, the layers will use available speed limits and the Speed Category for the rest.
How do I revert my Speed Limits to the Urban SDK default values?
A: Customers with Admin access can revert speed limits to their defaults within the Location Management Settings. Customers can choose to select a subset of links to revert or can select all links within their geography to reset all to their default values.
How is 50th Percentile Speed different than Average?
A: The 50th percentile speed, also called the median, represents the speed at which half of the drivers are traveling slower and half are traveling faster. It’s a middle point that is not influenced by extreme values like unusually high or low speeds. On the other hand, the average speed is calculated by adding up all the observed speeds and dividing by the number of observations. This means it can be skewed by outliers, such as a few drivers going extremely fast or slow. To learn more about the different data values and their definition check out the Traffic Speed Data Specification.
Why use 95th Percentile Speed as an alternative to Maximum Speed for safety prioritization and enforcement?
A: The 95th percentile speed is often used instead of the maximum because it’s a more reliable metric for enforcing speed limits. Maximum speed can be skewed by rare outliers, such as a single instance of extreme speeding, which doesn’t reflect typical driving patterns. The 95th percentile excludes the fastest 5% of speeds, giving a more accurate picture of the speeds most drivers are traveling at, which is critical for understanding whether there is a broad or isolated speeding trend that needs to be enforced. To learn more about the different data values and their definition check out the Traffic Speed Data Specification.
How can I identify the specific hour with the overall minimum and maximum speed?
A: Insights Reports show the overall minimum and maximum speeds for a selected period but do not specify the hour these occurred. To find the minimum or maximum speed for a specific day and hour, customers need to adjust the report filters to focus on that specific time period.
Other Data Questions
Can Urban SDK provide origin-destination (OD) data?
A: Urban SDK does not currently offer OD data due to technological limitations in achieving accurate trip data consistently across service markets. Urban SDK is developing a method to estimate trip distribution using traffic volume data and customer-defined screenlines. This will enable customers to estimate development trip distribution for traffic impact studies, calculate the impacts of road closures for special events or construction, and estimate pass-through traffic within a neighborhood, addressing the most commonly requested OD data needs.
Where does the Demographic and Socioeconomic data come from?
A: The US Census. We maintain an index of demographic, socioeconomic, economic and geographic boundaries necessary for long range transportation planning. We try to maintain 5 years historical information for each Dataset. However, the US Census periodically and without much notification may alter available attributes in various datasets. So, attributes in the same files may vary year over year.