Definitions

This section provides definitions and explanations for the key metrics, core statistical measures, commonly used performance and reliability indicators and concepts used throughout traffic studies, helping users understand how traffic performance, reliability, and safety metrics are calculated and interpreted.

This page covers the following definitions:

• Sample size
• Count estimations
• Road network length
• Speed and travel time
• Percentile
• Standard deviation
• Planning Time Index (PTI)
• Level of Service (LOS)
• Speed pace statistics
• Percentage over speed limit

Sample Size

The number of GPS devices observed on a specific road segment during the selected date range and time set.

The number shown in the results is always the total number of GPS devices we observed - we don't do any scaling, estimations, or averages when it comes to this value.

Total Sample Size

The total sample size shown in the study is the sum of all probe vehicle sample counts collected across: All days included in the Date Range (e.g., all weekdays in September 2025) within the selected Time Set (e.g., 6:00–6:15 each weekday).

Average Daily Sample Size

The total sample size divided by the number of days included in the selected date range represents the average number of probe vehicles observed per day within the specified time set.

Count Estimation

Average Daily Count Estimation (ML-Estimtaion)

The Average Daily Count Estimation represents the estimated average number of total vehicles per day on a road segment for the selected Date Range and Time Set.

Average Total Count Estimation (ML-Estimation)

The Average Total Count Estimation represents the estimated total number of vehicles traversing the road segment across the entire Date Range, within the selected Time Set.

Road Network Length

The total length of all the segments in your route or within your area. This value is used for pricing calculations. Segments with no data are not included in road network length calculations.

The number of date ranges multiplies the road network length. For example, if a 20 km route is analyzed over 5 date ranges, your verified road network length will be 100km.

When creating an Area-full or Area-density analysis, the road network length will be calculated by totaling the length of each selected FRC and then adding those lengths together.

Speed & Travel Time

Speed and travel time can be summarized using different statistical measures, each offering a distinct view of traffic conditions. The following will help you understand these statistical measures when reviewing your results.

Median Speed & Travel Time

A numeric value separating the higher half of a sample, a population, or a probability distribution from the lower half. The median of a finite list of numbers can be found by arranging all the observations from the lowest value to the highest value and picking the middle one. If there is an even number of observations, then there is no single middle value; the median is then usually defined to be the average, or mean, of the two middle values.

Harmonic Average, or Harmonic Mean

The harmonic average (or harmonic mean) is a way to calculate an average that is especially useful when dealing with rates, like speeds, where values are combined in a reciprocal manner. Unlike the arithmetic mean, which is a simple sum divided by the count, the harmonic average gives more weight to smaller numbers, which makes it particularly useful in situations where the relationship between values is not linear. The formula for the harmonic average of a set of numbers is:

Where:

• n is the total number of values.
• xi represents each individual value.

Arithmetic Average, or Arithmetic Mean

A mathematical representation of the typical value of a series of numbers is computed as the sum of all the numbers in the series divided by the count of all numbers in the series.

The formula for the arithmetic average or arithmetic mean of a set of numbers is:

Example

Imagine you have a car traveling along two segments of a route:

• First segment: The car travels 60 km at a speed of 80 km/h.
• Second segment: The car travels another 60 km, but at a speed of 40 km/h due to traffic.

The arithmetic average does not accurately reflect the true average speed over the entire journey because it doesn't consider the time spent at each speed.

We want to calculate the average speed over the entire 120 km journey.

• The car spends more time traveling at the slower speed of 40 km/h.
• The harmonic average takes into account the time spent traveling at each speed, providing a more realistic average.

Given the two speeds:

So, the car's true average speed over the entire journey is 53.33 km/h, not the arithmetic average of 60 km/h.

In situations where you're averaging rates (like speed), the harmonic average gives a better representation of the overall performance because it gives more weight to slower speeds, which are often more impactful in a journey.

Percentile

The value of a variable above which a certain percentage of observations fall. For example, the 20th percentile travel time is the value above which 20 percent of all the observations may be found. For example, a percentile of 50 is the median travel time for a route and a percentile of 90 means that that particular travel time has been achieved by at least 90% of all the GPS devices on the route.

Standard Deviation

Standard deviation is a widely used measure of variability or dispersion. It shows how much variation there is from the "average". A low standard deviation indicates that the data points tend to be very close to the average or mean, whereas a high standard deviation indicates that the data is spread out over a large range of values.

A slightly different explanation uses a normal distribution or bell-shaped curve. When the data samples are tightly bunched together and the bell-shaped curve is steep, the standard deviation is small. When the samples are spread apart and the bell curve is relatively flat, that tells you that you have a relatively large standard deviation.

Planning Time Index (PTI)

Planning Time Index is the ratio of the 95th percentile to the free-flow travel time and shows the total time that is needed for on-time arrival in 95 percent of all trips.

A PTI value of 2.0 for a given period suggests that travelers should spend twice as much time traveling as the free-flow travel time to reach their destination on-time 95 percent of the time. The planning time index is useful because it can be directly compared to the travel time index on similar numeric scales.

Different percentile values can be used instead of the 95th percentile. This value depends on your desired level of reliability. The lower percentile value results in lower reliability.

Level of Service (LOS)

Level of service scores for urban street segments are determined for motorized vehicles, pedestrians, bicycles, and transit, with the LOS score for motorized vehicles having different criteria and being more easily measured. The performance measures for motorized vehicles are travel speed for through vehicles and volume-to-capacity ratio for downstream through movement. LOS scores for pedestrian, bicycle, and transit modes are usually based on traveler perception, and can be calculated for specific links or segments. LOS scores for motorized vehicles on urban roads are defined as:

Speed Pace Statistics

The pace is defined as the continuous speed range (typically 10–15 km/h wide) that contains the highest percentage of vehicles.

Using the percentile table:

• A candidate pace range is tested by selecting a lower and upper speed bound (e.g., 38–52 km/h).
• The percentage of vehicles whose speeds fall within that range is computed using the percentile distribution.
• The range that captures the largest share of vehicles is selected as the pace range.

Once the pace range is defined, vehicles are categorized into three groups:

• Below Pace: Vehicles traveling slower than the pace range start
• In Pace: Vehicles traveling within the pace range
• Above Pace: Vehicles traveling faster than the pace range end

These percentages are derived directly from the cumulative speed percentiles:

• % Vehicles Below Pace: percentile at pace range start
• % Vehicles in Pace: percentile at pace range end − percentile at pace range start
• % Vehicles Above Pace: 100% − percentile at pace range end

Pace studies provide insight into:

• Speed variability and driver behavior
• Traffic flow stability
• Safety and enforcement considerations
• Corridor performance beyond simple averages

Percentage Over Speed Limit

When Speed Limit Violation is selected as the visualization mode, an additional metric, Percentage Over Speed Limit, becomes available. This metric shows the share of vehicles exceeding the posted speed limit across a range of thresholds (e.g., +5, +10, +15, +20, +25, +30, +35, +40 km/h or mph).

This allows you to evaluate not only whether speeding occurred, but how severe the violations were.

Example

• 20% of vehicles exceeded the limit by +5 km/h,
• 10% exceeded the limit by +10 km/h,
• No vehicles exceeded higher thresholds.

This metric provides a clear picture of speeding intensity and helps identify segments with a higher safety risk

Table of contents

• Definitions
• Sample Size
• Total Sample Size
• Average Daily Sample Size
• Count Estimation
• Average Daily Count Estimation (ML-Estimtaion)
• Average Total Count Estimation (ML-Estimation)
• Road Network Length
• Speed & Travel Time
• Median Speed & Travel Time
• Harmonic Average, or Harmonic Mean
• Arithmetic Average, or Arithmetic Mean
• Percentile
• Standard Deviation
• Planning Time Index (PTI)
• Level of Service (LOS)
• Speed Pace Statistics
• Percentage Over Speed Limit