Traffic Speed

Urban SDK provides data to stream-line and enhance traffic analysis, routing, and planning with live and historical data.

Traffic Speed Schema

The dataset is available in JSON, Shapefile, GEOJSON, and CSV file types.

Column Name	Description	Type	Example
`link_id`	Unique and persistent ID tied to urban sdk road network.	String	`16981048`
`timestamp`	15 minute intervals of time data requested	String	`2021-10-01 01:00:00`
`speed`	95th percentile mph speed in float value (precision two)	Float	`52.82`
`confidence`	Indicates the proportion of real time data included in the speed calculation. *See below for detailed calculation.	Float	`0.7`
`funclass_id`	Indicates classification of the roads depending on the speed, importance and connectivity.	Integer	`1`
`length`	Length of road segment in miles	Float	`0.0093`
`county_name`	The associated county name.	String	`Duval`
`state_name`	The associated state name.	String	`Florida`
`geometry`	Geometry that includes latitude and longitudes of row.	String	`MULTILINESTRING ((-81.59791 30.24124, -81.59801 30.24135))`

Detailed Calculations

Speed

The expected speed along the roadway; will not exceed the legal speed limit.

Confidence

The value of the confidence field indicates the proportion of real time data included in the speed calculation. It is a normalized value between 5 and 40 with the following meaning:

confidence = 5 indicates gap-filled speeds
10 < confidence <= 15 indicates suggestive speeds
20 < confidence <= 25 indicates highly suggestive speeds
30 < confidence <= 35 indicates confident speeds
confidence = 40 indicates highly confident speeds

This field can be used to identify whether the data for a location is derived from real time probe sources or historical information only.

The algorithm for calculating confidence factor is derived from the well-established statistical concept of confidence level, and uses the following general formula:
𝑧=Δ√𝑁−1max(𝜎𝑠𝑚𝑝, 𝜆)
Where N is the number of probe samples, σsmp is their standard deviation, and 𝜆 and Δ are some fixed constants based on how probe speed samples are obtained and processed. For example, probe speed samples are always integers, so their calculated standard deviation is higher than if they were more precise. The 𝜆 factor compensates for that. z is a confidence metric, which is then mapped into the ranges above.

Use this field in conjunction with your particular analytical case to control outliers. For example, if you are calculating an average behavior over a period of time, you could choose to omit records marked with the lowest confidence level, to reduce the effect of outliers.

funclass_id

The functional class is used to classify roads depending on the speed, importance and connectivity of the road.

Must satisfy: 1 ≤ value < 5

The value represents one of the five levels:

1: allowing for high volume, maximum speed traffic movement
2: allowing for high volume, high speed traffic movement
3: providing a high volume of traffic movement
4: providing for a high volume of traffic movement at moderate speeds between neighbourhoods
5: roads whose volume and traffic movement are below the level of any functional class

Traffic Speed

Contents:

Traffic Speed Schema