Title of Paper
Using GIS for Transportation Planning
Author’s Name
James R. Miller and Kevin G. Broecker
Abstract
The presentation will cover the process needed to integrate data from third party software into ArcGIS 8.1 for traffic signal optimization on a countywide basis. The purpose of this study was to determine where and when traffic signal timing needed to be adjusted to help optimize the flow of traffic for over 300 signalized intersections. The presentation will review the workflow for reformatting data from the Synchro Traffic Signal Timing software and making it useful in ArcGIS. The types of data collected, the database design, and how the data are analyzed to optimize traffic signals will be reviewed.
Paper Body
Purpose of Study - Hubbell, Roth & Clark, Inc. (HRC), was contracted by the Road Commission for Oakland County (RCOC) to perform a traffic signal optimization study for over 300 signalized intersections and cross-overs throughout the County. The Transportation Department at HRC was the lead consultant on this project which collected traffic count data for the intersections in the study area that approximately covered the area between Pontiac, Birmingham and Farmington, MI.
Background – Oakland County, Michigan, is located in southeast Michigan just north of Detroit and encompasses an area of over 900 square miles. With a 2000 population of almost 1.2 million residents, the land use of Oakland County is very diverse, ranging from high density communities with over 7000 people per square mile to rural farmland and large natural areas in the form of state and regional parks.(1) Oakland County is also characterized by a large number of lakes and waterways, which add to the aesthetic appeal of the area, but also poses a significant challenge in the area of transportation planning.
The many small lake communities have, over the past 30 years, coalesced into large sprawling residential communities, often times being services by the same two-lane roads, which have been restricted by historic housing development patterns around the lakes. To help maintain the quality of life and avoid significant economic impacts to the employers and residents of the County, the Road Commission for Oakland County has undertaken this project of optimizing the traffic signals throughout a large portion of the County to maximize the flow of traffic and minimize the need for expanded road construction projects. The RCOC, along with MDOT and several of the communities in the County, applied for and obtained funding from the Congestion Mitigation and Air Quality (CMAQ) for funding of this multi-year project.(2) CMAQ is a program under the EPA’s Clean Air Act intended to fund project that will help areas suffering from air pollution brought on by high auto emissions.(3) Most large urban centers in the country would qualify for these types of projects.
Technical Approach – HRC and the project team worked with the RCOC to identify the pre-timed and semi-actuated intersections throughout the study area in the County. HRC coordinated the project team to collect hourly traffic counts for the various approaches to each of the selected intersections and entered the data into the Synchro Traffic Count software database.
Synchro is a traffic signal optimization software package produce by Trafficware. They have approximately 1600 users throughout North America and are used by most state DOT’s. As an established traffic signal optimization software package, Synchro has a consistent database, designed to gather and analyze the necessary data for this type of a study.(4)
Each intersection is assigned a unique identifier (INTID) for data analysis and mapping purposes. The INTID is the key field which can be found in each of the data tables and provided the eventual link to the ArcGIS software. The data collected for this study was organized into one of five tables by the Synchro software. The five tables and a generalized description are as follows. A detailed list of the data fields and descriptions can be found in the Appendix.
Layout Table – Contains the x and y coordinates for map display.
Lanes Table – Contains the number of lanes for each direction of travel and turning configuration. Also contains the default settings for intersections with missing data or for entering projected future data.
Volumes Table – Contains the raw hourly traffic counts for each lane grouping and each direction of travel.
Timing Table – Contains the number of seconds for each phase of the intersection lights.
Phasing Table – Contains the adjustable timing setting for each phase of the intersection lights.
The Synchro software will by itself perform the traffic signal optimization and give the necessary results to adjust the timing and phasing at all the intersections for the study area. Synchro produces a schematic drawing of the intersection layout but does not relate to any other spatial data. In discussing the project with the Transportation Department staff, the GIS Department staff at HRC proposed to bring the data into ArcGIS to more accurately display the spatial component of the data and to take advantage of the additional analysis that could be performed in the GIS environment.
Integration with GIS – With the goals in mind of providing improved spatial accuracy, integration of additional data and the ability to provide further quality control, the GIS Department staff undertook the challenge of integrating the Synchro data into the ArcGIS environment.
The export format for the five Synchro data tables varied between .dxf files for the Layout Table and .utds or .dat files for the other four tables. The x and y coordinates from the Layout Table represent the graphical base for displaying the data from the other four tables, but it has a significant limitation in that any data displayed, such as traffic volumes, is displayed in the same layer as the street centerlines and intersection points. The display of the data is also limited in terms of the line and text types. The authors felt this would be a key advantage that the GIS software could offer in improving the legibility and analysis of the data.
Also of note is that the maximum coordinate limitations of Synchro are 200,000 units on the x and y values. This might impact large study areas such as a regional traffic analysis performed just within the Synchro software itself. This limitation also does not allow Synchro to directly import or export common coordinate data such as State Plane Coordinates.
The export format of the Layout Table, .dxf, is the native format for AutoCAD which ArcGIS is also able to read. The intersection coordinates were initially loaded into AutoCAD where they were reprojected into Michigan State Plane Coordinates and then combined with the centerlines from the GIS. This process can also be done using MicroStation as the CAD platform. The intersections were aligned with the positionally correct centerline intersections and given the new coordinate values from the GIS. AutoCAD can then export this new file as a .shp file.
The other four tables (Volume, Lanes, Timing, Phasing) need to be renamed from .dat files to .txt files so they can be read by MS Access. Once in an Access format the header information needs to be stripped off so the data can be read directly in the ArcGIS table.
With all five tables reformatted, the Layout.shp can be joined to any of the other four tables through the INTID key field. The attributes can now be viewed and analyzed with the various features of ArcGIS.
One of the challenges to producing hardcopy maps or displaying several attributes for the intersections at once was the fact that all the data wants to display on top of the intersection point. The authors made extensive use of the Text Label Settings, Text Label Symbols, and Label Placement Options to view up to 12 variables per intersection at one time. This much data was not practical for viewing or analysis from the Synchro software.
Conclusion – The three goals spelled out in the “Integration with GIS” phase of the project were all successfully completed. The first goal, providing improved spatial accuracy, was accomplished by underlying an orthorectified photo of the area obtained from the County GIS department. The major intersections were spaced reasonable well and with a slight rotation were brought into line with the orthophotos. Where the aerial proved especially useful was in locating the crossover intersections that are typically located before and after the major intersections. These crossover intersections are also signalized and, if out of sequence with the main intersection, can cause major delays and increased accident rates.
The second goal, integration of additional data, provided the most obvious advantage to this project. The GIS staff was able to integrate aerial photos, right-of-ways, parcels and parcel ownership, centerlines, bus stop locations and land use data from City, County and regional data sources. This provided the Transportation Department staff a much better perspective of the study area and allowed for the identification of an alternate corridor, in one example, to provide a smoother flow of traffic through one of the most congested areas of the County.
The third goal, the ability to provide further quality control, was shown to be useful when displaying data for a “T” intersection which was inadvertently assigned “through” traffic counts. In additional analysis in which the hourly traffic counts were put into a time lapse .avi file, another intersection was identified that had missing counts for one of the directions of travel. Identifying this by manually reviewing the data tables would have been very unlikely.
In summary, the advantages and disadvantages of integrating ArcGIS into this type of project are;
Pro’s
n Can use existing centerline files. This saves the expense of drawing a street network and provides a spatially accurate map that can overlay with other GIS data.
n Able to display a variety of other data such as the aerials, parcels, ROW’s, demographic data, construction plans, etc.
n Real world look & feel gives a wider audience the ability to draw conclusions
n Impact of “improvements” easier to evaluate. Multiple scenarios can be run by having the data in the GIS.
n Better public display can build concensus among policy makers and funding sources for improvement projects.
Con’s
n Additional learning curve on GIS software for the novice user.
n Additional expense of GIS software if an agency or consultant has not used this technology before.
n More data to keep track of can cause the need for expanded network resources.
n Increased metadata requirements will necessitate more staff to be properly trained in the creation and interpretation of metadata.
n Small projects may not see a positive cost/benefit. With the additional steps involved with bringing data into the GIS, a larger project may be required to see the benefit of the increased time and resource commitment.
Acknowledgements
The authors would like to acknowledge Mr. Keith McCormack for his support and encouragement of the GIS Department and its activities and Mr. Richard Beaubien and Mr. Prasad Nannapaneni of the HRC Transportation Department for their assistance and review of our procedures and analysis.
Appendix
The following are the field names for each of the five Synchro Tables.
Layout Table Volume Table Lanes Table Phasing Table Timing Table
INTID DATE Recordname Recordname PLANID
INTNAME TIME INTID INTID INTID
TYPE INTID Lanes BRP PMPEAK
X NBL Shared MinGreen S1
Y NBT Width MaxGreen S2
NID NBR Storage VehExt S3
SID SBL StLanes TimeBefore S4
EID
SBT
Grade
TimeToReduce S5
WID
SBR
SignControl MinGap
S6
NEID EBL IdealFlow Yellow S7
NWID EBT LostTime AllRed S8
SEID EBR SatFlow Recall CL
SWID WBL SatFlowPerm Walk OFF
NNAME WBT SatFlowRTOR DontWalk LD
SNAME WBR HeadwayFact PedCalls REF
ENAME NEL Volume MinSplit CLR
WNAME NET Peds
NENAME NER Bicycles
NWNAME NWL PHF
SENAME NWT Growth
SWNAME NWR HeavyVehicles
SEL BusStops
SET Midblock
SER Distance
SWL TravelTime
SWT
SWR
Definitions are for the Synchro Table Fields shown above. All definitions are taken from the Synchro Help menus.
Actuated Control- Traffic-actuated control of isolated intersections attempts to adjust green time continuously, and, in some cases, the sequence of phasing. These adjustments occur in accordance with real-time measures of traffic demand obtained from vehicle detectors placed on one or more of the approaches to the intersection. The full range of actuated control capabilities depends on the type of equipment employed and the operational requirements.
Detectors- Some of the more common detector definitions are defined below.
Actuation - The operative response of any type of detector (call).
Call. A registration of a demand for the right-of-way by traffic at a controller unit.
BRP – Barrier, Ring, Position. This is a part of the controller and the data in this field includes three digits which specify the barrier, ring and position within the barrier of each phase.
Calling Detector. A registration of a demand during red interval for right-of-way by traffic (vehicles or pedestrians) to a controller unit.
Check. An output from a controller unit that indicates the existence of unanswered call(s).
Continuous-Presence Mode. This is a mode of operation where the detector output continues if any vehicle (first or last remaining) remains in the zone of detection.
Controlled Output. This is the mode of operation where the detector has the ability to produce a pulse that has a predetermined duration regardless of the length of time a vehicle is in the zone of detection.
Detector. A device for indicating the presence or passage of vehicles.
Extension Detector. A detector that is arranged to register an actuation at the controller unit only during the green interval for that approach so as to extend the green time of the actuating vehicles.
Limited-Presence Mode. This is a mode of operation where the detector output continues for a limited period of time if vehicles remain in zone of detection.
Locking and Non-Locking Mode of Operation. Vehicle actuations (calls) can be received at the detector in either a locking or non-locking mode. For the locking mode, the call is retained until the phase receives it's green interval. For non-locking mode, the call is retained only while vehicles are in the zone of detection.
Passage Detection. The ability of a vehicle detector to detect the passage of a vehicle moving through the zone of detection and to ignore the presence of a vehicle stopped within the zone of detection.
Presence Detection. The ability of a vehicle detector to sense that a vehicle, whether moving or stopped, has appeared in its zone of detection.
Pulse Mode. This is a mode of operation where the detector produces a short output pulse when detection occurs.
Zone of Detection. The area or zone that a vehicle detector can detect a vehicle.
Saturation Flow Rate - Synchro 4 automatically calculates saturation flow rate for Right Turns on Red. This saturation flow rate is applied to a movement whenever the movement has a red signal. This calculation is also made for Left Turns on Red for crossing one-way streets.
The calculation of the RTOR Saturation Flow Rate is quite complex and is based on the signal timing, the volumes of the subject approach, and the volumes of any merging approaches.
Grade- The grade is the slope for traffic approaching the intersection. Use a negative grade for downhill.
Growth Factor- The growth factor can be used to adjust the traffic volumes. The raw volume data is multiplied by the growth factor when calculating Adjusted volumes and Lane Group volumes The growth factor can be 0.5 to 3.0.
Current volumes can be adjusted to future volumes with the Growth Factor. To calculate a growth factor based on a growth rate over several years, use the following formula.
GF = (1+r)^Y = Growth Factor
r = Growth Rate
Y = Number of years
For example the growth factor for 3% growth over 10 years is:
GF = (1 + 0.03) ^ 10 = 1.34
Heavy Vehicles- Enter the percentage of vehicles that are trucks or buses for this movement. This value affects the saturated flow rate shown in the LANE window. The default for this field is 2%.
Heavy Vehicle percentages affect input flows, but not turning percentages in CORSIM.
Headway Factor- The Headway Factor is based on the lane width factor, the grade factor, the parking factor, the bus stops factor, and the area factor. The Headway factor is magnified by 30% because at cruise speeds, about 30% of the time per vehicle is taken by vehicle passage and 70% by the headways.
Ideal Flow- Enter the ideal saturated flow rate for a single lane in this field. The 1997 HCM recommends using 1,900 vehicles per hour per lane. This is the default.
Lost Time- The Total Lost Time is the amount of time lost for a phase change. It includes Startup Lost time plus the Unusable clearance time. This value should be at least 3 seconds. All effective green times are taken as the Split minus the Total Lost Time. Startup Lost Time is usually taken as 2.0 seconds, this is the time it takes vehicles to get going when the light turns green. Unusable Clearance Time is the total Clearance Time minus the time that vehicles continue to enter the intersection during clearance. It is usually 1 second or more. For longer yellow and all-red intervals, this value may be higher.
MaxGreen- The maximum green time for each phase in seconds.
MinGap- This is the minimum gap time that the controller will use with volume-density operation. If volume-density operation is not used, set this value to the same as the Vehicle Extension.
MinGreen- This field is the minimum initial green time for a phase. This is the shortest time that the phase can show green. A typical value would be 4 seconds. The minimum value allowed in Synchro is 1 second.
MinSplit- The Minimum Split is the shortest amount of time allowed for this phase.
The minimum split must at least be long enough to accommodate the Minimum Initial interval plus the yellow and all red time. This will usually be 8 seconds or more.
PH or PHF- The traffic volumes are divided by the Peak Hour Factor (PHF) to determine the traffic flow rate during the busiest 15-minute period during the hour. For example:
Hourly Flow Rate: 1000 vph
Peak Hour Factor: 0.9
Adjusted Peak Flow Rate: 1000 / 0.9 = 1111 vph
If you have 15-minute counts, you can use the highest counts and a PHF of one.
Phase numbers- the labels assigned to the individual movements around the intersection. For an eight phase dual ring controller, it is common to assign the main street through movements as phases 2 and 6. Also, it is common to use odd numbers for left turn signals and the even numbers for through signals. A rule of thumb is that the sum of the through movement and the adjacent left turn is equal to seven or eleven.
Recall- Each phase can have a recall of None, Minimum, Maximum, or Ped.
No Recall: The phase can be skipped.
Minimum Recall: The phase will always come on to its minimum, the phase can not be skipped.
Maximum Recall: The phase will always show its maximum and has no detection. The phase cannot skip or gap out, nor can it be extended.
Pedestrian Recall: The phase will always show a walk phase. The phase can't be skipped or gap out until the walk and don't walk intervals have passed.
Coord Recall: Used with coordinated signals only. This phase shows for its maximum plus any unused time from preceding phases.
PedCalls- This is the number of pedestrian push button calls for this phase. This value is only needed if this phase has a pedestrian push button.
Shared Lanes- lanes that serve more than one movement
Sign Control- If Control Type is set to Unsignalized, this intersection becomes unsignalized and the third row is Sign Control.
There are three options:
1. Free: traffic goes through the intersection without stopping,
2. Yield: traffic has a yield sign and slows down, stopping only if necessary.
3. Stop: all traffic stops, and waits until all conflicting traffic is clear.
Storage Lanes- the number of lanes in the right or left storage bay. This value only appears when the storage length is greater than 0. By default the number of storage lanes is equal to the number of turning lanes.
TimeBefore- When using volume-density operation, this is the amount of time before gap reduction begins.
TimeToReduce- When using volume-density operation, this is the amount of time to reduce the gap from Vehicle Extension (or maximum gap) to Minimum Gap.
VehExt- The vehicle extension (or gap) is the unit time extension for each vehicle actuation.
References
(1)
(2) Road Commission for Oakland County, Request for Proposal, February, 2001
(3) Federal Highway Administration web site, http://www.fhwa.dot.gov/environment/cmaqpgs/index.htm
(4) Trafficware web site and personal interview. http://www.trafficware.com/synchro/index.html
Author Information:
Jim Miller
Senior GIS Specialist
Hubbell, Roth & Clark, Inc.
555 Hulet
Drive
Bloomfield Hills, MI 48302
Phone: 248-338-9241 x238
Fax:
248-454-6312
Email: jmiller@hrc-engr.com
Kevin Broecker
CADD/GIS Specialist
Hubbell, Roth & Clark, Inc.
555 Hulet Drive
Bloomfield Hills, MI 48302
Phone: 248-338-9241 x208
Fax: 248-454-6312
Email:
kbroecker@hrc-engr.com