Bottleneck Mapping

Summary

Bottleneck Mapping is a subproject of the Mobility Performance Report, which is one of the products of the Mobility Performance Reporting and Analysis Program (MPRAP). The Mobility Performance Report is prepared by the California Department of Transportation (Caltrans) and District staff to provide detailed data about highway system performance related to congestion and mobility. Caltrans collects vehicle counts and calculates speeds at all hours of the day and all days of the week in major metropolitan areas throughout California via the Caltrans Performance Measurement System (PeMS--see Data Source tab). This information helps identify congestion bottlenecks and results in more cost-effective investments to improve the performance of the State Highway System.

Background

PeMS defines a bottleneck as “a persistent and significant drop in speed between two locations on a freeway.” Bottlenecks are determined by the bottleneck identification algorithm in PeMS. This algorithm looks at speeds along a facility and declares a bottleneck at a location where there has been a drop in speed of at least 20 mph between the current detector and the detector immediately downstream. This speed drop must persist for at least five out of any seven contiguous five-minute data points, and the speed at the detector in question must be below 40 mph. While PeMS identifies the detector locations where these conditions are met, these bottleneck locations are only approximate (based on the locations where detectors are present).

The bottlenecks identified through the PeMS Bottleneck Identification Algorithm are filtered by a number of factors to obtain the bottlenecks mapped in the documents below. This filtering was done to create a consistent bottleneck analysis process for all districts, and to only report bottlenecks that are recurrent and causing large amounts of delay. The bottlenecks reported include bottleneck locations that were active on at least 20 percent of all weekdays during the year, persisted for at least 15 minutes on average, and caused more than 100 vehicle hours of delay (VHD) per weekday. This filtering means that some rural districts had less than 10 bottlenecks to report in the AM Peak and PM Peak periods. These top bottleneck locations are shown on District PDF maps, along with lines depicting the congestion queue length resulting from these bottlenecks. If a District had more than 10 bottlenecks that met the criteria described above, those additional bottlenecks and their congestion queues are shown on the District’s map.

Process

Bottleneck maps are created by the GIS staff in the Assets Management Branch of the Office of Performance, Division of Traffic Operations, using data obtained by each district from PeMS (the Performance Measurement System--see the Data Source tab). The data are processed and filtered in an Excel workbook. MPRAP staff review all district submissions and clear them for mapping. GIS staff use ArcGIS to convert the processed data into KML (Keyhole Markup Language) files that can be viewed at the street level in Google or Caltrans Earth. District technical personnel review their KML files and ground truth (use their engineering judgment and local knowledge to validate) each congestion segment and verify the bottleneck locations and lengths. The revised data are then exported into PDFs and JPEGs that are inserted into the annual Mobility Performance Report. In some cases, districts revise their workbooks, or request changes based on the appearance of the PDFs, and the whole process is repeated until the districts have validated all bottlenecks, and all map versions match.

For further information on Bottleneck maps, please contact Huilan Han.