Chapter 15

ADVANCED TRAFFIC MANAGEMENT SYSTEM

DATA

Robert L. Bertini
Department of Civil and Environmental Engineering, Portland State University

Ahmed El-Geneidy
School of Urban Studies and Planning, Portland State University

With the implementation of Intelligent Transportation Systems (ITS) for

system management purposes, there is now the ability to extract archived data
that can be used to evaluate the implementation of new operational strategies.
In recognition of the need to provide feedback to decision-makers, efforts are
underway to provide rigorous documentation of ITS benefits and costs. The
objective of this paper is to describe how Advanced Traffic Management
System (ATMS) data are being used to contribute toward these evaluations.
Case examples are described in the areas of freeway management, incident
management, arterial management, and transit management. Building a
complete ITS system requires collaboration in time, funding, and institutional
arrangements. ITS components that are integrated can result in synergistic
effects when considered as an entire system. It is shown that in some cases it
is possible to build upon national level statistics describing ITS benefits by
using data collected from the systems themselves. It is hoped that further
efforts to integrate transportation planning with evaluation methodologies will
incorporate the necessary empirical results from a wide variety of studies. In
this way, better databases can be developed, and heightened accountability
will be more pervasive in the evaluation of ITS improvements.

1. INTRODUCTION

In 1997 in the U.S., automobiles traveled 1.4 trillion vehicle miles (2.3
trillion vehicle kilometers) and households spent an average of 19 percent of
their income on transportation—less than housing but more than food
288 Chapter 15
(Northeast-Midwest Institute, 2002). Further, drivers in the 68 largest urban
areas in the U.S. experienced an increase in traffic delays due to congestion
from 11 hours per year in 1982, to 36 hours per year in 1999 (Schrank and
Lomax, 2002). The estimated cost of traffic congestion in these 68 areas
totaled $78 billion, representing a cost of 4.5 million extra hours of travel and
6.8 billion gallons (25.7 billion liters) of wasted fuel (Schrank and Lomax,
2002). The average rush-hour trip takes 32 percent more time than the same
trip taken during non-rush-hour conditions. Congested travel periods (rush
hours) in the nation’s major cities have doubled in less than 20 years,
increasing from nearly three hours (morning and evening combined) in 1982,
to almost six hours in 1999 (Schrank and Lomax, 2002). Congestion is now
found during almost half of the daylight hours on workdays (Schrank and
Lomax, 2002).
Increasing traffic congestion coupled with improved technology, funding
constraints, and increasing environmental consciousness has provided an
impetus to develop cost effective systems aimed at improving the efficiency
and effectiveness of the transportation system. Intelligent Transportation
Systems (ITS) include a wide range of diverse technologies, including
information processing, communications, control, and electronics. ITS have
evolved with applications, including collision warning systems, ramp meters,
advanced signal control systems, transit and emergency vehicle management
systems, and others. The goals of ITS deployments include improving traveler
safety, traveler mobility and system efficiency; increasing the productivity of
transportation providers; and conserving energy while protecting the
environment. The strain on the transportation system as a whole is thus eased
through the application of modern information technology and
communications. Some technologies provide more cost-effective benefits than
others, and as technology evolves, the choices to deployers are bound to
improve. These technologies are often combined into a single integrated
system, providing benefits that exceed the benefits of any single technology
(Proper and Maccubbin, 2000).
ITS aims to improve the safety and efficiency of the transportation
system. ITS systems themselves offer opportunities for new methods of
evaluation and continuing assessment. As an indication of the degree of
commitment to ITS in the U.S., during the last decade, federal, state, and local
governments have appropriated billions of dollars for ITS programs. In 1998,
the Transportation Equity Act for the 21st Century (TEA-21) provided more
than $1.2 billion in funding to support ITS through 2003. Of that, $603
million was targeted toward research and development. Another $679 million
was intended for deployment of ITS projects (Sundeen, 2002). Further, the
Intelligent Transportation Society of America (ITS America) estimates that
more than $209 billion will be spent on ITS programs by 2011.
Advanced Traffic Management System Data 289
There has been recognition of the need to demonstrate the benefits of ITS,
providing a necessary feedback loop to decision-makers. In order to facilitate
evaluations of ITS investments, the USDOT through its ITS Joint Program
Office (established in 1994) continues to collect information regarding the
impacts of ITS projects on the operation of the surface transportation network.
The results of most ITS related projects and model deployments have been
perceived as promising and efforts continue toward defining the magnitude of
their benefits (Proper and Maccubbin, 2000). The objective of this paper is to
describe how advanced traffic management system (ATMS) data are being
used to evaluate the benefits of ITS investments.

2. EVALUATION PERSPECTIVES

Transportation improvements have frequently been deployed in order to

make the system more efficient by reducing travel time, the number of stops,
and delay. In the past, data collection experiments would be designed for
limited time periods to collect data at the precise points of interest. This
process was often manual and costly, with many people required to collect a
small amount of data. Bias was then introduced by temporally extrapolating
these data collected from one or two days over an entire year. The data that
were used to estimate the benefits of the improvement had limited temporal
coverage but were collected from the precise spatial points of interest.
Many ITS deployments include surveillance systems that are used in the
operation of the system. For example, ramp metering systems usually include
inductive loop detectors on the freeway and on-ramps in order to detect the
presence and traffic state of vehicles. This information is often used to set the
ramp meter timing, but from an evaluation standpoint, it is not possible to
“choose” the loop detector locations, and it is impossible to move the
detectors. Despite the fact that the detectors may have been located
specifically to operate the metering system, it is recognized that the
surveillance system can also be used for incident detection and verification. It
has also been shown (Bertini, Leal, and Lovell, 2001; Nee, Ishimaru, and
Hallenbeck, 2001; Ishimaru, Hallenbeck, and Nee 2001; Ishimaru and
Hallenbeck, 1999) that these systems can be used to extract relevant
performance characteristics for the transportation network and then tracked
over time. In contrast to past evaluation efforts where an experiment could be
designed and data could be sampled from “ideal” locations for short periods,
ITS deployments have provided opportunities to sample data from “non-
ideal” locations (established for traffic management purposes) over very long
periods. The bias that results originates from the need to extrapolate over
space, rather than over time, since data can be collected indefinitely.
290 Chapter 15
3. ITS BENEFITS OVERVIEW

Since December of 1994, the USDOT’s Joint Program Office (JPO) for
ITS has collected information describing the impact of ITS projects on the
operation of the surface transportation system. Data collected as part of these
efforts are available in the ITS Benefits Database on the JPO Web site
(www.its.dot.gov). The JPO also collects information on ITS costs, and
maintains this information in the ITS Unit Costs Database. The database is a
central site for estimates of ITS costs data that the JPO can use for policy
analyses and benefit-cost analyses. In addition, the database can be viewed
and downloaded as a cost-estimating tool for those implementing ITS projects
and programs at state and local levels.
The development and deployment of ITS technologies offer a wide
variety of opportunities for local, regional, and state agencies to improve the
capacity, reliability, and efficiency of their transportation systems. Due to
many factors, the quantification of ITS benefits and costs has been difficult
using traditional transportation planning and analysis methods because
traditional transportation planning models lack necessary sensitivity to many
benefits derived from ITS technologies, and because information on the
impacts and costs of many ITS technologies is not yet well-understood.
The Federal Highway Administration (FHWA) and others have
recognized this potential barrier to integrating ITS into the transportation
planning process. In 1997, FHWA and its partners began development of the
ITS Deployment Analysis System (IDAS), which is a tool designed to help
planners better address these issues (Cambridge Systematics, 2002).
Cambridge Systematics, Inc. led the development team and the software is
now available for use. According to product documentation, IDAS is designed
to assist public agencies and consultants in integrating ITS in the
transportation planning process. IDAS offers the capability for a systematic
assessment of ITS with one analysis tool and is used for determining the
benefits and costs of various ITS deployments. IDAS provides users with the
following capabilities:
Comparison and screening of ITS alternatives;
Estimation of impacts and traveler responses to ITS;
Estimation of life-cycle costs;
Inventory of ITS equipment, and identification of cost-sharing
opportunities;
Sensitivity and risk analysis;
ITS deployment and operations/maintenance scheduling; and,
Documentation for transition into design and implementation.
As with any model, IDAS is not without limitation. IDAS operates as a
post-processor for travel demand model output, and incorporates benefit and
Advanced Traffic Management System Data 291
cost information from many disparate studies, with data coming from
different locations and from different timeframes. Thus it should be
emphasized that IDAS is only a tool and should be used with care. More
research is needed to quantify actual ITS deployment benefits, and the results
of such research should be incorporated into future ITS evaluation activities.

4. ITS COMPONENTS

ITS deployments themselves typically include surveillance systems that

enable a more comprehensive understanding of how the existing
transportation system operates and facilitates proactive strategies for
managing it more efficiently. ITS deployments have benefited from advances
in computer processing and miniaturization, communications technology, and
enhanced institutional arrangements. Ten ITS systems will be introduced with
some examples of how archived data have been used to provide evidence for
the effectiveness of these systems. The ITS benefits and unit costs database
has classified the benefits of implementing ITS into the following 10 program
areas (USDOT, 2002a). Note that each program area includes different ITS
applications and that there is some potential overlap:
Freeway management
Incident management
Transit management
Arterial management
Emergency management
Electronic payment
Traveler information
Crash prevention and safety
Operations and maintenance
Road weather management.
The ITS benefits and unit cost database also describes seven categories of
benefits to be used for ITS deployment evaluations:
Safety improvements
Delay savings
Throughput
Customer satisfaction
Cost savings
Environmental
Other.
In line with the scope of this chapter we concentrate on the benefits by
program area and use some real examples used in the evaluation process. The
292 Chapter 15
examples are derived from past and ongoing efforts to evaluate specific
systems using archived ATMS data.

5. FREEWAY MANAGEMENT SYSTEMS

Three primary ITS functions make up freeway management systems:

monitoring and surveillance, control of freeway operations, and the display or
provision of information to motorists via dynamic message signs, highway
advisory radio, in-vehicle navigation or information systems, or specialized
information transmitted only to specific set of vehicles. Evaluations of
freeway management system improvements such as ramp metering systems
have demonstrated improvements in safety, reduction in travel time and delay,
increased flows, and flow improvements (USDOT, 2002b).
Despite early efforts to deploy metering and management systems, actual
traffic monitoring over a widespread area and real-time response is easier now
due to advances in technology and greater system coverage. Typical traffic
operations centers (TOCs) collect and process surveillance and monitoring
data, most often from inductive loop detectors, and supplemented this with
closed circuit television (CCTV) cameras that are also directly controlled
from the TOC. The ability to collect data and reflect on it in real time has
made a difference.

5.1 Interstate 5 Evaluations in Portland, Oregon

Presently, projects are underway by the authors to evaluate the

performance of ramp metering and incident management in the Interstate 5
corridor in Portland, Oregon. The Oregon Department of Transportation
(ODOT) has deployed a systemwide ramp metering program throughout the
Portland metropolitan area. As part of the system itself, approximately 400
inductive loop detectors have been installed. Loops are included on each
metered on ramp and in the freeway mainline lanes just upstream of each
metered ramp. However, there are no detectors on off-ramps and there are few
detectors at intermediate points (between interchanges) on the freeways.
These detectors report speed, count, and occupancy every 20 seconds, but
ODOT only archives data aggregated over 15-minute periods. Through
special arrangement for the evaluation projects, the raw data are being
archived. Thus far, the loop detector data has been validated with vehicle
count data manually extracted from the video surveillance system. Figure 1
shows two sample validation curves, which include cumulative vehicle
arrivals (plotted on oblique axes in order to magnify the details). The curves
extracted from the loop detectors are aligned with the curves extracted from
the surveillance video, indicating that the detectors are functioning reliably.
Advanced Traffic Management System Data 293
294 Chapter 15
The ramp metering evaluation will not include a true “before” and “after”
component since ODOT is not able to shut down the meters. The project will
include a comparison of actual ramp and freeway corridor performance using
actual detailed counts and speeds extracted from the raw loop detector data.
Using simulation tools, attempts will be made to evaluate different metering
strategies and to create hypothetical performance characteristics “without”
metering. It is recognized that relying on the fixed loop detector locations,
with potentially large spacing, some bias will result in the analysis due to
assumptions made about speeds over long freeway sections. As an example,
Figure 3 shows a map of a section of Interstate 5 describing corridor speeds
measured by loop detectors. When computing corridor performance, each
detector is assigned to an influence area bounded by the midpoints between
detectors. If the head or tail of a queue is present in the section, speed/travel
time computation errors will be introduced which will in turn affect the
calculation of performance measures such as vehicle miles (kilometers)
traveled, delay, vehicle hours traveled, etc. Using a probe vehicle (at
approximately the same time), Figure 2 shows the speed profile experienced
by an actual vehicle traversing the section. Experiments are underway to
examine the benefits of fusing the loop detector data with probe vehicle
information (incident response vehicles and express buses are equipped with
automatic vehicle location systems).
The incident management evaluation underway relies on the automatic
vehicle location (AVL) system data provided by the incident response
vehicles themselves, as well as an archived incident database that includes
input from the incident responders, TOC dispatchers, and other emergency
vehicle personnel.
Advanced Traffic Management System Data 295
296 Chapter 15
5.2 Minnesota Ramp Metering System Evaluation

Recent public opposition threatened to abandon ramp control as a traffic

management option in the Twin Cities of Minneapolis and St. Paul. In
response to this, the Minnesota DOT was asked to produce tangible
independent evidence of the effectiveness of ramp metering. A data collection
procedure started with the ramp meters in operation, and continued with the
meters shut down. A comparative study was conducted to compare before and
after shutdown data (Cambridge Systematics, Inc., 2001; Levinson et al.,
2002). In order to identify the temporal and spatial extents of congestion,
occupancy contour plots were used (Bertini, Leal, and Lovell, 2002). Figure 4
shows two plots of occupancy with and without ramp meters along a 16-mile
section of Minnesota Trunk Highway 169 (TH-169). Note the increase in
occupancy (corresponding to increased travel times across the loop detectors)
when the ramp meters were turned off compared to the occupancies measured
with ramps turned on during another day. The before analysis indicated
congestion somewhere between stations 17 and 18 while less congestion
between stations 22 and 23 was reported in the after plot. Tables 1 and 2
summarize performance characteristics for a portion of TH-169. As shown,
the speed dropped by nearly 18 percent after the shut down of the ramp
metering system. Table 2 shows that the vehicle miles traveled (VMT)
decreased by approximately 9.4 percent after the metering system was shut
down.
The analysis performed by Cambridge Systematics, Inc. (2001) indicated
that ramp metering is a cost-effective investment for the Twin Cities area,
finding that after the meters were turned off, there was an average nine
percent traffic volume reduction on freeways and no significant traffic volume
change on parallel arterials included in the study. During peak traffic
conditions, freeway mainline throughput declined by 14 percent in the un-
metered condition. It was also estimated that the ramp metering contributed to
an annual savings of more than 1,000 crashes or approximately four crashes
per day. From an environmental perspective, ramp metering results in a net
annual savings of 1,160 tons (1,052 metric tons) of emissions.
In parallel to the above study, a microsimulation analysis (Hourdakis and
Michalopoulos, 2002) computed nearly the same benefits gained from
applying ramp metering technology. The main result of the simulation process
was that it has developed a prototype validated by empirical analysis under
the relatively unique circumstances of a shutdown. See Chapter 9 of this book
by Zhang and Levinson for another evaluation of the Twin Cities ramp
metering shut-off.
Advanced Traffic Management System Data 297

Table 1. Minnesota TH-169 Average Speed

Before Shutdown After Shut Down Percentage Decrease

Speed (mph) Speed (mph) Speed (mph)
Stations Cumulative Average Cumulative Average Cumulative Average
16 190,000 65 130,000 48 32 26
17 170,000 57 140,000 47 18 18
18 210,000 75 185,000 65 12 14
19 185,000 66 145,000 53 22 20
20 180,000 63 160,000 56 11 11
21 180,000 62 150,000 52 17 16
22 190,000 67 160,000 55 16 20
Average Speed Reduction from Station 16 to 22 = 18%

Table 2. Minnesota TH-169 Vehicle Miles Traveled

VMT VMT VMT % Difference

Veh-miles Veh-miles Veh-miles VMT
Station Miles Before After Before - After Before - After
15 0.098 4637 4183 453 9.8
16 0.17 11478 10355 1122 9.8
17 0.26 13387 12027 1359 10.2

18 0.37 18667 16760 1907 10.2

19 0.50 24065 21699 2366 9.8

20 0.45 22381 20312 2069 9.2

21 0.42 19436 17770 1665 8.6

22 0.45 20214 18560 1653 8.2
Total 34268 21670 12597 9.4
298 Chapter 15
Advanced Traffic Management System Data 299

5.3 Incident Management Systems

Incidents are defined as crashes, breakdowns, and other random events

that occur on our highway system. Congestion caused by incidents are serious
problems that face any transportation agency. Incidents are known to cause
more than 50 percent of urban congestion and lead to economic losses, air
pollution, and human pain and suffering. Many urban areas have developed
quick response incident management systems, recognizing that transporting
victims to trauma centers within the “golden hour” can save lives. Further,
through coordination among highway operations, law enforcement, and
emergency personnel, secondary crashes can be prevented and responder
safety can be enhanced (El-Geneidy and Bertini, 2003). So incident
management systems are coordinated, preplanned, and/or real-time use of
human resources to reduce the duration of incidents (Zografos,
Androutsopoulos, and Vasilakis, 2001). Incident management systems contain
components such as incident detection, incident verification, response to the
incidents, clearance of the incidents, and traffic management at the incident
locations. In many locations, incident data are archived on a regular basis to
identify locations of high incident frequency. These locations can be used in
planning the responders’ routes on the highway and for identification of
reasons for incident causation in an effort to improve the existing roadway
characteristics to avoid future incidents at the same location. Numerous
studies have been conducted to evaluate the implementation of incident
management programs. Most of these studies came to the same findings that
incident management programs have a substantial effect on delay time.
(Chapter 11 of this book by Parthasarathi, Levinson, and Gillen also examines
driver’s willingness to pay for freeway service patrols).

5.4 Bay Area Freeway Service Patrol Evaluation

Incident management programs have been very popular additions to the

transportation system. In many ways, the Bay Area Freeway Service Patrol
Evaluation (Skabardonis et al., 1995) set the standard for comprehensive
evaluations of incident management systems since it was a true before-and-
after analysis. The study collected 276 hours of field data on one nine-mile
(14.5 kilometer) freeway section, for 24 weekdays before and 22 weekdays
after the freeway service patrol (FSP) was implemented. Field data included
incident observations, probe vehicle travel times and speeds, flows, and
occupancies extracted from archived loop detector data. The loop detectors
were spaced at approximately 1/3 mile (0.5 kilometer) increments on the
freeway mainline (and on-ramps) and the probe vehicles traveled at seven-
300 Chapter 15
minute headways during peak periods. Based on estimated savings in incident
delay and fuel consumption, the study found that the FSP was cost effective
for that particular freeway segment. It may have been tempting to try to
extrapolate the results of the analysis to other freeway sections in the Bay
Area or in other area. However, the researchers emphasized that the results
would only be applicable “to sites with traffic and incident characteristics
similar to the ones in the study area” (Skabardonis et al, 1995).

5.5 Los Angeles Freeway Service Patrol Evaluation

The Los Angeles Freeway Service Patrol Evaluation (Skabardonis et al.,

1998), a true “before-and-after” analysis, measured the effectiveness of the
FSP on a 7.8-mile (12.6 kilometer) section of the I-10 freeway in Los
Angeles. An evaluation methodology was developed to estimate incident
delays based on field data from loop detectors and probe vehicles, and to
derive estimates of savings in performance measures in the absence of data
for “before” FSP conditions. The methodology required the application of
response time (and thus incident duration) savings due to the presence of the
FSP on the freeway. Field data were collected in the corridor for 32
weekdays, for a total of six hours each day. This 192-hour database includes
detailed descriptions for 1,560 incidents, probe vehicle travel time runs at 5.7-
minute headways, and archived flow, occupancy and speed data from 240
loop detectors. This study found that the FSP program was also effective for
the corridor studied.

5.6 Oregon Region 2 Incident Response Evaluation

In Oregon, an evaluation of the rural Region 2 Incident Response (IR)

program (Bertini et al., 2001) involved a statistical analysis of archived
incident data, estimation of reductions in fuel consumption and delay,
calculation of program costs, and development of a decision-making tool for
design/expansion of the incident management program on future corridors.
The methodology consisted of a quantitative analysis of archived incident
data during two distinct phases since the IR program’s inception. The study
focused on a 51-mile (82 kilometer) corridor on Oregon Highway 18 and a
41-mile (66 kilometer) corridor on Interstate 5 in Lane County. As shown in
Figure 5, Phase 1 covered the period between February 1995–March 1997 and
Phase 2 covered the period from March 1997–December 2000. Figure 5 also
shows the total amount of IR resources deployed. As shown, during Phase 1
on Highway 18, IR personnel invested approximately 36 hours per month,
while during Phase 2 (continuing today), there is one full-time IR staff
member deployed (173 hours per month). Figure 5 also indicates that the
Advanced Traffic Management System Data 301
staffing level has increased over time on Interstate 5. It was not possible to
conduct a true “before-and-after” study, since it is the IR staff themselves who
are the roving data collectors. Thus, the numbers of reported incidents (rather
than the actual number of incidents) has increased because the IR personnel
are physically monitoring the status of the roadways. Figure 6 is a tree that
displays the number of incidents observed during the two phases. This
illustrates that potential bias that can be introduced when using a data
collection system that relies on the incident response personnel themselves.
302 Chapter 15
Advanced Traffic Management System Data 303
5.7 Transit Management Systems

Transit management systems are concerned with increasing operational

efficiency of all transit modes and increasing ridership by making the transit
system more reliable. (McQueen and McQueen, 1999) The emergence of the
global positioning systems (GPS) and the increase in its accuracy has helped
this field substantially. Several transit agencies have equipped their vehicles
with GPS to create automatic vehicle location (AVL). AVL technology has
been widely implemented in North America and Europe. In the year 2000
about 35 bus systems had AVL technology implemented in the U.S., in both
light-rail and bus systems (APTA, 2001).
The Tri-County Metropolitan Transportation District of Oregon (TriMet)
operates 97 bus routes and a 38-mile light rail line within the tri-county
Portland metropolitan region. TriMet’s bus lines carry approximately 200,000
trips per day, serving a total population of 1.3 million persons within an area
of 590 square miles (1,530 square kilometers). TriMet is considered as one of
the leading ITS deployers in the U.S. TriMet has implemented a Bus Dispatch
System (BDS) as a part of its overall operation and monitoring control
system. (Strathman et al., 2002; Strathman et al., 2000; Strathman et al., 1999)
The main components of the BDS include:
Automatic vehicle location (AVL) based upon differential global
positioning system (GPS) technology, supplemented by dead
reckoning sensors;
Voice and data communication system using radio and cellular digital
packed data (CDPD) networks;
On-board computer and control head displaying schedule adherence
information to operators, detection, and reporting of schedule and
route adherence to dispatchers;
Automatic passenger counters (APCs) on front and rear doors of most
vehicles (Kimpel, 2001); and
Computer-aided dispatch (CAD) center (Strathman et al., 2001).
The implementation of the BDS in Portland, Oregon has resulted in
substantial savings to the existing system and increased the service reliability
in the region for both bus and light-rail (Strathman, et al., 2002). The total
annual benefits derived from implementing the TriMet BDS system is
estimated at $5.4 million dollars, and the present value imposing a 12-year
expected life on the BDS is $47.8 million.

5.8 Arterial Management Systems

An arterial management system is used to manage traffic by employing

various detection and control devices along arterial roadways. This includes
304 Chapter 15
surveillance and traffic signal control, and sometimes includes audio or visual
information on arterial roadway conditions. Detectors collect basic traffic
condition data (typically flow and speed information) and adaptive control
systems can be used to coordinate traffic signal control across a metropolitan
area by adjusting the lengths of signal phases and cycles. Without centralized
control, vehicles would be delayed at intersections irrespective of actual
traffic conditions as the vehicle progressed through the route. This caused
undue vehicular delay to all vehicles including transit vehicles. Using
knowledge of real-time traffic characteristics and coordination, arterial
management systems have contributed to reductions in red light violations of
20–75 percent and reductions in fuel consumption by 2–13 percent in the
studied areas (USDOT, 2002a). It was shown that St. Paul, Minnesota, traffic
signal preemption systems reduced crashes for emergency vehicles by 71
percent in seven years (USDOT, 2002a). An arterial management system can
be also monitored by the existing vehicles running on the system. For
example several transit agencies have equipped their vehicles with GPS which
reports the location of the vehicle back to a dispatch center every few
seconds.
As an example, using some of TriMet archived data, a fusing process was
developed as shown in Figure 7. This figure displays time-space diagrams for
both TriMet buses and “ground truth” probe vehicles that were traveling on
Powell Boulevard at the same time on the same day. The bus AVL data was
extracted from archived data for Route 9 beginning at Front Avenue on the
west side to the intersection of Powell Boulevard and SE 39th Ave. The probe
vehicle data were collected with GPS installed on the vehicle. The bus AVL
data were archived through BDS described above. A comparison between
travel time for both the bus and the vehicle is shown in Table 3. The
integration between these two systems can also be compared using vector
analysis in Geographic Information Systems (GIS). Figure 8 shows an
interpolated surface comparing a bus speed “surface” with a “surface”
constructed from probe vehicle speeds. The surface was created using 20 runs
collected by the GPS installed on the probe vehicles. The surface was
interpolated using a krigging method in the ArcGIS software. Another vector
surface was created for 20 bus trips using the same method and during the
same period of time. Both the probe vehicles and buses had the same origin
and destination. Looking in depth at the comparison, the bus has behaved in
the same way as the probe vehicle yet the percentage of drop in speed at some
locations were not similar. These differences are due to the variations in speed
between the two modes during free flow travel time. Statistical relations
between the probe vehicle and bus performance can be developed for
reporting arterial performance to travelers and to quantify improvements to
arterial management systems.
Advanced Traffic Management System Data 305
306 Chapter 15
5.9 Emergency Management Systems

Emergency management systems are used by fire departments, police

departments, ambulance services, and freeway service patrols. These systems
respond to emergencies and direct the various departments to the incident
location through the shortest path in order to clear the incident or save a life.
These systems include traffic signal priority to give the right of way to the
departments’ vehicle. The emergency management systems use the AVL
technology in order to locate the nearest vehicle and direct it to the incident.
This system is managed by a transportation management center. The delivery
of emergency service to the communities is an important responsibility that
should be met when any person is facing an emergency.
A study of the Minnesota Highway Helper Program found that the
program reduced the duration of a stall by eight minutes. Based upon
representative numbers, annual benefits through reduced delay totaled $1.4
million for a program that cost $600,000 to operate. While in another pilot
study looking at the Courtesy Patrol Program in Denver, Colorado, the
estimates concluded a reduction cost in traffic delay by $0.8–$1.0 million for
the morning period and by $0.90–$0.95 million in the evening. The study
assumed a time value of $10 per hour. Program costs varied between the tow
truck operators between $29 to $38 per truck-hour, which results in a benefit-
to-cost ratio of 10.5:1 to 16.9:1 (USDOT, 2002a).
Advanced Traffic Management System Data 307

5.10 Electronic Payment

Electronic payment systems are present on many of the highways in the

U.S. Several DOTs are turning to toll collection in order to finance new roads
and maintain existing highways. The congestion caused upstream of toll
booths began to be a problem so the idea of electronic payments has emerged
as an important response. Typically, drivers subscribe to an electronic
payment system and are given radio frequency (RF) transponders that
communicate with the toll collection system. Vehicles passing through the toll
facility entrance and/or exit are not required to stop as their payment is
automatically deducted from their accounts (Klein, 2001). Electronic payment
is also used for collecting transit fares and commercial vehicle operating fees
where the transponder can be used in various ways and it is linked to a bank
account or credit card line. A typical manual toll lane might process 350
308 Chapter 15
vehicles per hour while applying electronic payment on all lanes will result in
about 1,200 vehicles per hour (ITE, 2000). If the toll plaza was eliminated the
rate could be 2,000 vehicles per hour per lane. This application will save on
both toll booth construction and administration fees.

5.11 Traveler Information

Traveler information systems are used to inform travelers regarding road

conditions via broadcast media. The system collects data regarding the current
status of the transportation network and broadcasts it to travelers via
communication channels and media (McQueen et al., 1999) The objective is
to provide travelers with current information so they can avoid congested
routes. This kind of system tries to avoid the externalities caused by additional
vehicles in the congested system. The communication system can be one-way
or two-way where the vehicle will be equipped with GPS to identify the
vehicle location and a traveler information center would direct the vehicle to
an uncongested route. This system is known as a vehicle-motorist service
information system (Hulse, Dingus, and Barfield, 1997). Several DOTs have
started to apply similar systems and have begun to broadcast one-way
communication to travelers via the radio and via the Internet. In Seattle,
Washington, as shown in Figure 9, a Web-based traveler information system
is available on the Internet (Washington Department of Transportation, 2002).
The system is updated every minute. This kind of information system can also
be implemented for transit. Several transit agencies have implemented
Internet-based trip planners to transit riders. These trip planners save time and
increase reliability to transit services.

5.12 Crash Prevention and Safety

In 1990, there were an estimated 16 million U.S. vehicle crashes. Forty-

five thousand fatalities occurred during these crashes, along with 5.4 million
nonfatal injuries and 28 million damaged vehicles. The average cost per crash
is approximately $8,600. Crashes are mainly caused by human errors,
including errors in recognition, decision, and performance. An ITS-based
crash prevention and safety system will include an advisory crash avoidance
system to alert the driver with a warning when the vehicle detects a crash is
about to occur. This system can include an advisory system to indicate the
optimum headway and the best speed (Dingus, et al., 1997). Vehicles can also
be equipped with an in-vehicle safety and warning system where warnings of
immediate hazards and road conditions affecting the roadway ahead of the
driver are reported to the driver.
Advanced Traffic Management System Data 309
5.13 Operations and Maintenance

Operations and maintenance systems are created during the process of

implementation of any ITS application to measure the success or the decline
of the system. Operations and management systems are encouraged by the
USDOT. The USDOT is responsible for monitoring 75 metropolitan areas in
the U.S. that have deployed ITS and received federal funding for ITS
investments. In addition, archived surveillance and performance data can be
used later for generating various performance measures (Bertini and El-
Geneidy, 2003; Bertini, Leal, and Lovell, 2002) and feeding performance data
back into the planning process. Performance measures can lead to a better
understanding of the existing system and the archived data can be used by
various stakeholders in ways we cannot yet imagine. Collection of every
single type of data needed for advanced traffic control for the entire traffic
system is unrealistically costly and inefficient. Processing of disorderly and
incomplete information reported from the field is usually complex and time
consuming. (Klein, Yi, and Teng, 2002).
310 Chapter 15
Advanced Traffic Management System Data 311
5.14 Road Weather Management

Weather impacts on transportation are pervasive. The weather can cause

many incidents especially in the cold regions of the country. A study trying to
quantify the benefits of an anti-icing program in seven different states in the
United States was conducted in order to encourage the use of anti-icing/road
weather information system technologies. The strategy of anti-icing involves
the use of chemical freeze point depressants to prevent a bond from forming
between pavement and snow or ice. NCHRP Project 20-7, Task 117 was
initiated to address these needs and quantify the benefits (Boselly, 2001). The
study concluded that the anti-icing program can reduce costs of providing a
defined level of service by 10–20 percent, while the snow and ice control
costs per lane mile can be reduced up to 50 percent.

6. CONCLUSIONS

After considering the 10 different ITS component systems it is clear that

each system cannot be deployed to stand alone in the overall transportation
system. Building a complete ITS system requires collaboration in time,
funding, and institutional arrangements. ITS components that are integrated
can result in synergistic effects when considered as an entire system. It is
shown that in some cases it is possible to build upon national level statistics
describing ITS benefits by using data collected from the systems themselves.
Thus far, the quantification of ITS benefits has not been statistically
sophisticated. Often, benefits are expressed as being “certain,” when this is far
from the truth. For example, the measurement of any reduction in mean travel
time as a result of an ITS deployment involves bias; thus any benefit should
be expressed along with its associated variance. In addition, there is no
guarantee that travel time reduction due to the installation of ramp metering in
one city will result in similar benefits in another city—particularly if the
nature of system integration and institutional cooperation is widely different.
It is hoped that further efforts to integrate transportation planning with
evaluation tools such as IDAS and microsimulation will incorporate the
necessary empirical results from a wide variety of studies. In this way, better
databases can be developed, and heightened accountability will be more
pervasive in the evaluation of ITS improvements.
312 Chapter 15
ACKNOWLEDGEMENTS

The authors are indebted to numerous individuals for the provision of data
and assistance in preparing this paper. Oregon Department of Transportation
(ODOT) video and loop detector data were provided by Jack Marchant and
Dennis Mitchell of ODOT Region 1. TriMet data were generously provided
by Steve Callas of TriMet and Tom Kimpel and Prof. James Strathman of
Portland State University. Sutti Tantiyanugulchai assisted with providing the
Powell Blvd. data. Prof. David Levinson, University of Minnesota, kindly
provided data from the ramp meter shutdown study and Shazia Malik of
Portland State University assisted with processing the Minnesota data and
gathering important background information.

REFERENCES
American Public Transit Association. APTA Vehicle Data Book, Washington, DC, 2001.
Bertini, R., Tantiyanugulchai, S., Anderson, E., Lindgren, R., Leal, M. “Evaluation of Region 2
Incident Response Program Using Archived Data,” Portland State University,
Transportation Research Group, Research Report, 2001.
Bertini, R. L., El-Geneidy, A. M. “Using Archived Data to Generate Transit Performance
Measures,” Washington DC, Transportation Research Board, Annual Meeting.
Bertini, R. L., Leal, M., Lovell, D. J. “Generating Performance Measures from Portland’s
Archived Advanced Traffic Management System Data,” Washington DC, Transportation
Research Board, Annual Meeting, 2002.
Boselly, E. “Benefit/Cost Study of RWIS and Anti-icing Technologies” (20-7(117)),
Washington DC, National Cooperative Highway Research Program, Transportation
Research Board, National Research Council, 2001.
Bureau of Transportation Statistics. Omnibus Travel Survey, 2000.
Cambridge Systematics, Inc. “Twin Cities Ramp Meter Evaluation Technical Report,”
Minneapolis, MN, Minnesota Department of Transportation, 2001.
Cambridge Systematics, Inc. “Intelligent Transportation Systems Deployment Analysis
System,” Software Documentation, 2002. Retrieved November 19, 2002, from
http://idas.camsys.com.
Dingus, T. A., Jahns, S. K., Horowitz, A. D., Knipling, R. “Human Factors Design Issues in
Crash Avoidance Systems,” Human Factors in Intelligent Transportation Systems, W.
Barfield and T. Dingus eds. Mahwah, NJ: Lawrence Erlbaum Association, Inc., 1997.
El-Geneidy, A. M., Bertini, R. L. “Integrating Geographic Information Systems and Intelligent
Transportation Systems to Improve Incident Management and Life Safety,” the 8th
International Conference on Computers in Urban Planning and Urban Management
Conference, Sendai, Japan, 2003.
Evanco, W. M. “A Data Fusion Framework for Meta-evaluation of Intelligent Transportation
System Effectiveness” (Report 0495 18B40A), MITRE working note, Center for
Information Systems, 1996. Retrieved November 19, 2002, from
http://plan2op.fhwa.dot.gov/pdfs/Pdf1/Edl04454.pdf
Hourdakis, J., Michalopoulos P. G. “Evaluation of Ramp Control Effectiveness in Two Twin
Cities Freeways,” Washington DC, Transportation Research Board, Annual Meeting, 2002
Advanced Traffic Management System Data 313
Hulse, M. C., Dingus, T. A., Barfield, W. “Description and Application of Advanced Traveler
Information Systems,” Human Factors in Intelligent Transportation Systems, W. Barfield
and T. Dingus eds. Mahwah, NJ: Lawrence Erlbaum Association, Inc, 1997.
Institute of Transportation Engineers. Intelligent Transportation Primer, Washington, DC,
Institute of Transportation Engineers, 2000.
Ishimaru, J., Hallenbeck, M. E. “FLOW Evaluation Design Technical Report,” Washington
State Transportation Center (TRAC), 1999.
Ishimaru, J., Hallenbeck, M. E., Nee, J. “Central Puget Sound Freeway Network Usage and
Performance,” Washington State Transportation Center (TRAC), 2001, 1999 update.
Klein, L. A. Sensor Technologies and Data Requirements for ITS, Norwood, MA: Artech
House (Intelligent transportation systems library), 2001.
Klein, L., Yi, P., Teng, H. “Decision Support System for Advanced Traffic Management
through Data Fusion and Mining,” Washington, DC, Transportation Research Board,
Annual Meeting, 2002.
Levinson, D., Zhang, L., Das, S., Sheikh, A. “Ramp Meters on Trial: Evidence from the Twin
Cities Ramp Meters Shut-off,” Washington DC, Transportation Research Board, Annual
Meeting, 2002.
McQueen, B., McQueen, J. Intelligent Transportation Systems Architecture, Norwood, MA:
Artech House (Intelligent transportation systems library), 1999.
Nee, J., Ishimaru, J., Hallenbeck, M. E. “HOV Lane Performance Monitoring: 2000 Report,”
Washington State Transportation Center (TRAC), 2001.
Northeast-Midwest Institute. “Roads and Highways,” 2002. Retrieved November, 18, 2002
from http://www.nemw.org/roadshighways.htm.
Peng, Z., Beimborn, E., Neluheni M. “A Framework for the Evaluation of the Benefits of
Intelligent Transportation Systems” (Report). Milwaukee, WI, Center for Urban
Transportation Studies, University of Wisconsin Milwaukee, 2000 Retrieved November
19, 2002, from http://www.uwm.edu/Dept/CUTS/its/itsb121.pdf.
Proper, A. T., Maccubbin, R. “ITS Benefits: Data Needs Update 2000,” 2000. Prepared in
connection with the 12th July ITS Benefits Data Needs Workshop, Mitretek Systems.
Schrank, D., Lomax, T. “The 2002 Urban Mobility Report,” Texas Transportation Institute,
Texas A&M University System, 2002. Retrieved November 18, 2002 from
http://mobility.tamu.edu.
Skabardonis, A., Noeimi, H., Petty, K., Rydzewski, D., Varaiya, P., Al-Deek, H. “Freeway
Service Patrol Evaluation,” University of California at Berkeley, California, PATH
Research Report, UCB-ITS-PRR-95-5, 1995.
Skabardonis, A., Petty, K., Varaiya, P. P., Bertini, R. L. “The Los Angeles Freeway Service
Patrol Evaluation,” University of California at Berkeley, California PATH Research
Report, UCB-ITS-PRR-98-31, 1998.
Strathman, J. G., Kimpel, T. J., Dueker, K. J., Gerhart, R. L., Turner, K., Griffin, D., Callas, S.
“Bus Transit Operation Control: Review and an Experiment Involving TriMet’s
Automated Bus Dispatching System” (Paper 01-0438), Washington, DC, Transportation
Research Board, Annual Meeting, 2002.
Strathman, J. G., Dueker, K. J., Kimpel, T., Gerhart, R., Turner, K., Taylor, P., Callas, S.,
Griffin, D., Hopper, J. “Automated Bus Dispatching, Operations Control, and Service
Reliability,” Transportation Research Record, 1666, 1999, 28–36.
314 Chapter 15
Strathman, J. G., Kimpel, T. J., Dueker, K. J., Gerhart, R. L., Turner, K., Taylor, G., Griffin, D.
“Service Reliability Impacts of Computer-aided Dispatching and Automatic Vehicle
Location Technology: A TriMet Case Study,” Transportation Quarterly, Vol. 54, No. 3,
2000, 85–102.
Sundeen, M. “The Expanding Role of Intelligent Transportation Systems,” National
Conference of State Legislatures, 2002. Retrieved November 18, 2002, from
http://www.ncsl.org/programs/esnr/ITStranrev02.htm.
United States Department of Transportation (USDOT). “ITS Benefits and Unit Cost Database,
2002a.” Retrieved November 18, 2002, from http://www.benefitcost.its.dot.gov/
its/benecost.nsf/ByLink/deskReference
United States Department of Transportation (USDOT). “Measuring ITS Deployment and
Integration” (Electronic Document Number (4372)), U.S. Department of Transportation,
Joint Program Office for Intelligent Transportation Systems, 2002b. Retrieved November
18, 2002.
Washington State Department of Transportation. “Puget Sound Traffic Conditions,” 2002.
Retrieved November 18, 2002, from http://www.wsdot.wa.gov/PugetSoundTraffic/.
Zografos, K. G., Androutsopoulos, K. N., Vasilakis, G. M. “A Real-time Decision Support
System for Roadway Network Incident Response Logistics,” Transportation Research
Part C, Vol. 10, 2002, 1–18.