Determining traffic congestion patterns
Embodiments generally relate to determining traffic congestion patterns. In some embodiments, a method includes identifying congestion events for each road of a plurality of roads in a road network, where each congestion event indicates a drop in average vehicle speed below a predetermined speed threshold for a particular road in the road network, and where the congestion events span a predetermined time period. The method further includes determining local clusters of the congestion events based on one or more road condition parameters, where each local cluster defines a local congestion pattern for a particular road of the plurality of roads in the road network. The method further includes grouping the local clusters into one or more global clusters based on the one or more road condition parameters, where the global clusters define global congestion patterns in the road network.
1. A computer-implemented method comprising:
identifying a plurality of congestion events for each road of a plurality of roads in a road network, wherein each given congestion event indicates a drop in average vehicle speed below a predetermined speed threshold for a particular road in the road network, and wherein the plurality of congestion events span a predetermined time period;
determining local clusters of the congestion events for the particular road based using a clustering method on one or more road condition parameters wherein each local cluster defines a set of local congestion patterns for the particular road of the plurality of roads in the road network; and
clustering, using a clustering method, respective ones of the local clusters into a plurality of global clusters, wherein a global cluster consists of a set of local clusters each having a similar congestion pattern, the set of local clusters for a plurality of respective ones of the plurality of roads and the global cluster is clustered independent of any spatial parameter.
2. The method of claim 1 , wherein the at least one processor further performs operations comprising identifying each road in the road network based on a road location in the road network.
3. The method of claim 1 , wherein the one or more road condition parameters further comprise traffic parameters.
4. The method of claim 1 , wherein the one or more road conditions further comprise weather parameters.
5. The method of claim 1 , wherein each congestion event is a regular congestion event.
6. The method of claim 1 , wherein the determining local clusters of the congestion events uses a spatial parameter and the clustering of the local clusters into a plurality of global clusters does not use a spatial parameter.
7. The method of claim 1 , wherein the determining local clusters of the congestion events uses only the events associated with a particular road are included in each local cluster and the clustering of the local clusters into a plurality of global clusters assigns different ones of the plurality of roads in the road network to a same global cluster.
8. The method of claim 1 , wherein a first global cluster contains a first local cluster belonging to a first particular road and a first local cluster belonging to a second particular road and a second global cluster contains a second local cluster belonging to a first particular road and a second local cluster belonging to a second particular road.
9. The method of claim 1 , wherein a first global cluster contains local clusters representing a first congestion pattern in the entire road network including a first local cluster belonging to a first particular road and a first local cluster belonging to a second particular road and a second global cluster contains local clusters representing a second congestion pattern in the entire road network including a second local cluster belonging to a first particular road and a second local cluster belonging to a second particular road.
10. The method of claim 1 , wherein the identifying of each congestion event comprises:
identifying a stable congestion seed within the predetermined time period where the average vehicle speed is below the predetermined speed threshold for the particular road in the road network;
searching a database of traffic data backward in time from the stable congestion seed to determine a start time of the given congestion event and forward in time from the start time to determine an end time of the given congestion event, the start time established when the average vehicle speed decreases below the predetermined speed threshold and the end time established when the average vehicle speed increases above the predetermined speed threshold; and
defining the given congestion event as a process of traffic congestion for a certain period of time from the start time to the end time, wherein the certain period of time varies between the plurality of congestion events.
11. The method of claim 1 , wherein the method further comprises determining one or more road condition parameter values for each congestion event, wherein the one or more road condition parameter values include a vehicle queuing length for the particular road during the certain period of time.