Route tracking

Route tracking analyzes data from vehicles or individuals to reveal patterns around your city.

Route tracking analysis

You might take all the route tracking data available and plot it on a map of your city, but you would probably have a map that displays traffic on every street. By further analyzing the data to determine most popular routes, you can get a better idea of the typical routes. Route tracking analysis takes into account all of the data points available, calculates route segments, and aggregates routes that match specified criteria to display aggregated route data on the map. The routes are displayed with colors that indicate popularity (a ranking of frequency).

Each route segment that is evaluated must belong to the same vehicle, and the time interval and distance of the segment must be within a certain threshold (maximum duration and maximum distance).

The available data is evaluated to identify clusters of data points, and route segments are identified between two clusters. Route segments are matched to municipal street data to ensure validity. The segments are then aggregated to determine the most common route between the two data point clusters. These aggregated routes are displayed on the map. An aggregated route might not represent any of the individual routes that are taken, but are close to actual routes.

The route tracking analysis further takes into account factors such as time, distance, and speed to filter out obvious errors.

Display details

Distance: Distance in miles or kilometers from starting to ending data points for the selected track segment.
Frequency: The frequency of a segment is the number of times that vehicles traveled this segment in a particular time range. The same vehicle that travels the same route repeatedly increases the frequency.
Popularity: The popularity of a segment is based on the frequency. All route segments are sorted by frequency, then assigned popularity score. The top 10% most frequently traveled segments receive a popularity score of 4. A popularity score of 3 goes to the top 10% to 30% frequency segments. A popularity score of 2 goes to the top 30% to 60% frequency segments. And popularity of 1 for the remaining segments.
Travel time: The average time that is required to travel the selected segment.

Analysis tuning parameters and default values

You can change the values for the route tracking analysis parameters in the Analysis Tuning page.

Graph ratio: A ratio that determines which and how many of the most traveled route tracks are displayed on the map. Popularity is assigned only to the tracks to be displayed.; For example, if the analysis returns a total of 100 route tracks and the Graph ratio is set to 0.6, only 60 of the 100 tracks are displayed on the map. Popularity values of 1 to 5 are assigned to only those 60 tracks.; The default value is 1.
Incidents per location: The minimum number of instances of tracks that coincide with each data point required to include that data point in the clustering analysis.; Generally, the target number of data points is between 3 percent and 10 percent of the total number of points in your data set. Incidents per location defines the minimum number of coincidental data points required to produce meaningful results from the route tracking analysis.; The default value is 3.
K-means clustering: A process that is used to partition a group of data points into a few clusters. K is the value that, when divided into the number of input points as determined by Incidents per location, returns 10 - 50 input points per cluster. K can vary depending on how many input points define meaningful clusters from your data.; For example, if your data includes 5,000 input points, as determined by Incidents per location, setting K = 200 returns 25 input points per cluster.; The default value is 100.
Maximum duration: The maximum time between two data points for the points to be considered a valid route segment. If the time between the two points exceeds the maximum duration, the segment is divided into two or more segments.; The default value is 1200 seconds (20 minutes).
Maximum speed: The maximum speed over a segment that is considered to be valid and not an error.; For example, if the data indicates that a particular segment was traveled at a speed of 200 miles per hour, that data is rejected as invalid.; The default value is 200 miles per hour or kilometers per hour.
Maximum distance: The maximum length of a segment, in miles, that is considered valid and not an error.; For example, a segment length of 2 miles might be considered too long for one track.; The default value is 0.6 miles or kilometers.
Heat map maximum distance: The minimum distance, in miles, between two data points to be considered too close together for evaluation as an individual track.; Depending on the size of the region being analyzed, Heat map maximum distance might be between 0.1 and 1 mile.; The default value is 0.2 miles or kilometers.