Automated Enforcement for Speeding and Red Light Running (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

30 City of Portland, Oregon, Automated Enforcement Program Overview The City of Portland, Oregon’s automated enforcement program started in January, 1996, as a demonstration project to test the effectiveness of photo radar as a speed enforcement tool. The red light camera program followed suit in October, 2001. Portland’s program, operating now for almost 15 years, has been very successful with a firm foundation in the public trust. They have also been committed to operating a transparent program that is marked by their availability to the news media. The following document presents an overview of their program and illuminates some of the best practices present that can be replicated by other agencies in the development and operation of an automated enforcement program. The information is based on interviews with key person- nel associated with Portland’s programs and published reports. Background Problem Identification The citizens of Portland are concerned about the livability of their community. The transpor- tation system is structured to accommodate all modes, including motor vehicles, pedestrians, bicycles, various transit vehicles, and other non-motorized users (Figure H-1). In the mid 1990s, speeding in neighborhoods was identified as a safety concern by both the city and neighborhood associations. The Portland Bureau of Transportation (PBOT) had been working with neighborhood groups to implement traffic calming measures. However, these measures were not enough. The Bureau identified automated speed enforcement as a potential tool to reduce speeding in neighbor- hoods. This was vetted with the neighborhood associations in what can be best described as grassroots campaigning and community building. The city has 95 neighborhood associations across seven districts. Bureau staff met directly with the neighborhood associations and initiated discussions about the potential benefits of this tool as a complement to the traffic calming efforts. The neighborhood associations overwhelmingly embraced this tool. Enabling legislation did not exist in Oregon in the mid-1990s. The Bureau encouraged each neighborhood association to write a letter to the legislature voicing their support for the adop- tion of enabling legislation for automated enforcement. The overwhelming support voiced at the grassroots level by these neighborhoods prompted the legislature to develop and pass enabling legislation. Case Studies A P P E N D I X H

Case Studies 31 Enabling Legislation In 1995, the City of Portland received authority from the Legislature to conduct a 2-year trial of photo radar enforcement. The legislature required certain elements to be completed as part of the trial period, along with regulations on how the program can be operated. Specifically, the legislation required Portland to 1. Provide a public information campaign to inform local drivers about the use of photo radar before citations are issued. 2. Conduct a process and outcome evaluation of the demonstration, for the Department of Trans- portation, that includes the effect of the project on traffic safety, the degree of public acceptance of the project, the process of administration of the project, and suggestions for design or plan- ning changes that might reduce traffic congestion on residential streets or use of such streets as thoroughfares. 3. Implementation requirements: a. Shall be confined to streets in residential areas or school zones. b. Shall be used no more than 4 hours per day in any one location. c. The photo radar equipment is operated by a uniformed police officer out of a marked police vehicle. d. An indication of the actual speed of the vehicle is displayed within 150 feet of the location of the photo radar unit. e. Signs indicating that speeds are enforced by photo radar are posted, so far as is practicable, on all major routes entering the jurisdiction. f. The citation is mailed to the registered owner of the vehicle within 6 business days of the alleged violation. g. The registered owner is given 30 days from the date the citation is mailed to respond to the citation (1). Figure H-1. Bicyclists in Portland, Oregon (Source: Greg Raisman).

32 Automated Enforcement for Speeding and Red Light Running The legislation allowed for a pilot program, restricted to certain communities in Oregon, including Portland. After completing a successful trial phase, the Legislature extended the use of photo radar under Oregon Revised Statutes (ORS) 810.438 and 810.439. These statutes address photo radar authorization, evaluation, and citations. The red light camera legislation is enabled under ORS 810.434, 810.435, and 810.436. These statutes are on the operations and evaluation of red light photos, the use of photography, and citations based on red light photos. The most recent legislations just removed the 12-camera limit that was required under the statutes. The restrictive legislation has helped with public acceptance of the program (2). Initiation of Enforcement With the legislative backing, Portland initiated a pilot automated speed enforcement program in January 1996. The Bureau had a dedicated full-time staff to start up the program. Mr. Robert Burchfield, the City Traffic Engineer, noted that in the initiation of the program he thought about the process, not just the outcome (3). At the forefront of his mind was the public trust. He wanted to ensure that a process was developed for the program that not only satisfied the legisla- tion, but also was transparent for the public. The Portland program demonstrated success and the pilot restriction was lifted. The remain- der of this document discusses the current program. Speed Enforcement Program Program Administration and Structure The Portland Police Bureau’s Traffic Division leads the speed enforcement program. They work closely with the currently contracted vendor (Affiliated Computer Services [ACS]) and the Portland Bureau of Transportation (PBOT). The program is a driver liability program. The vio- lations are issued to the registered owner of the vehicle if a good quality photograph is captured and the gender of the observed driver matches the registered owner. Operations Portland currently operates four photo radar vans—two full-time vans and two part-time vans. Two vans operate using film and two use digital photography. One benefit to the digital photography is that it allows the officer to identify any problems with the photos (e.g., glare on the windshield) and fix the problem in the field; with film there could be problems with the positioning of the van that will not be noticed until the film is developed. The vans are operated and run by the Portland Police Bureau. Figures H-2 through H-4 show outside and inside views of Portland’s digital photo radar van. There are certain restrictions that guide the operations of the photo radar program. Captain Nelson, who started the program, thought the more restrictive the program, the better the public opinion of it (2). The vans can only operate a maximum of four consecutive hours at one loca- tion during a deployment. Deployment is restricted to residential streets, construction zones, and school zones. The vans must be operated by a sworn police officer. At the site the officer must display a portable sign within 100 to 400 yards of the van that warns drivers of the photo radar enforcement ahead. Figure H-5 shows the portable sign used by the Portland Police Bureau. There is no time of day restriction for photo radar deployment. Site Selection Through City Ordinance #172517, the Portland Police Bureau is directed to use the photo radar vans in school zones, highway work zones, residential streets, and SAFE (Strategic and

Case Studies 33 Figure H-2. Portland’s digital photo radar van. Figure H-3. Rear view of Portland’s digital photo radar van. Focused Enforcement) zones. SAFE zones are areas within the jurisdiction that have been identi- fied as having a high number of speeding violations and speed-related crashes. Currently, there are 18 SAFE zones in the City of Portland (4). The SAFE zone sites are selected by PBOT and vetted with the Police. Sites that the public requests for speed enforcement are first observed by the police to deter- mine the best way to alleviate the speeding problem. The photo radar vans are rotated through- out the city between the SAFE zones and areas with complaints. Each site has a unique four-digit code that personnel use to monitor coverage. On average, the Police Bureau deploys the vans at 26,000 sites per year (2). They also generally do one deployment per day in a school zone. Warning Periods There was a warning period when the program started in 1996, along with an information campaign to notify and educate the public (2). An additional warning period is not used when

34 Automated Enforcement for Speeding and Red Light Running Figure H-4. Computer and camera monitor inside digital photo radar van. Figure H-5. Portable sign placed in advance of photo radar van.

Case Studies 35 new locations are added since the program is well established in the community and advance signing is provided. Citation Process There are five steps involved with issuing photo radar citations. 1. Violation detection. 2. Violation processing. 3. Quality control checks. 4. Citation review and approval. 5. Citation mailing. Speeding violations are detected when the police officer operating the photo radar vehicle visually observes a violation. The speed tolerance is 11 mph over the speed limit. In some cases, such as inclement weather, the tolerance will be lowered. The vehicle’s speed is displayed in the van, and the officer keeps a log and takes notes on each violation. When a violation occurs, a minimum of three photographs are generated, which include the vehicle approaching the van, a close up of the driver in the vehicle, and a close up of the vehicle’s license plate (4). The film in the cameras from the photo radar vans are developed by the vendor, ACS. ACS uses the photos to identify the license plate of the violating vehicle, and then requests the vehicle registration from the Oregon Department of Motor Vehicles (DMV). The vendor reviews the details of the violations, such as location, date, time, and speed. Violations are gender matched, which means violations are discarded when the gender of the driver does not match the gender of the registered owner of the vehicle. Violations are also thrown out if an identification cannot be made due to glare on the windshield, face blocked by a visor, etc. The vendor is responsible for the quality control checks of the citations. The photos are viewed by multiple people before being sent to the Oregon DMV for registration information. After the DMV sends the registration, the photos are viewed again and verified against the registration for gender match. The citations then go into an electronic queue on a secure Website where the police officer can log on and approve the citations. The citations from photo radar can be approved in batches by the police rather than individually. After the citations are approved, ACS conducts one more review for accuracy before mailing them to the vehicle owner (5). After the quality control checks, the citation is mailed to the registered owner within 6 busi- ness days. Included with the citation is a photo of the offending driver and a Certificate of Innocence. If the DMV has a flag on the registration, a Certificate of Innocence is not included with the citation. The registered owner has 30 days to respond in which they can either pay the violation fee or, if they were not the violating driver, file a Certificate of Innocence (5). The citation issuance rate for the photo radar is about 60% for film, and 75% for digital. Fac- tors that help increase the issuance rate are proactive maintenance of the equipment and proper training of the police officers. The most common factors for non-issuance are gender match and glare on the windshield (5). Police Department Role The Portland Police Bureau is the lead agency for the photo radar program. Their primary responsibility is in the operation and deployment of the photo radar vans. They currently have two full-time photo radar officers and 13 other trained officers. They have found that officers that are usually assigned to motorcycle duty are good candidates for this duty. The operation of the vans provides a safer duty for motorcycle officers during inclement weather. The Police Bureau’s goal is to have 200 hours per month of deployment, although with recent budget cuts the numbers have been reduced to 100 to 140 hours per month.

36 Automated Enforcement for Speeding and Red Light Running The Police Bureau is also responsible for the vendor selection. The vendor is selected through a competitive bid process. The Police Bureau is the liaison with the vendor. This involves daily communication and regularly scheduled weekly meetings. The Police Bureau also maintains frequent communication with the Bureau of Transportation and the city’s elected leaders. Vendor Role ACS is responsible for the daily processing of the photos and citations, from developing/ reviewing the film and collecting registration information to conducting the quality assurance/ quality control and mailing the citations. They also have a call center to respond to questions and comments from the public. The vendor works directly with both the Portland Police Bureau and the Portland Bureau of Transportation and is in contact with both Bureaus’ on a daily basis. ACS is also responsible for providing training courses for the police officers. The contract period is now 5 years with the vendor. Originally the contract period was shorter, allowing for more flexibility. However, now that the program is well established, the longer con- tract period is useful since it is a big effort to re-compete the contract. The Police Bureau noted that the contract should outline in detail the requirements of the vendor. For example, if a media blitz is needed, this should be outlined in the vendor’s contract. Adjudication A Certificate of Innocence is included in each citation mailing. If the registered owner of the vehicle was not the driver at the time of the violation, they must fill out the Certificate of Innocence and include a copy of their license. If the photograph in the citation does not match the photograph on the license, the citation is thrown out. However, if the photographs match, a second citation is mailed out, this time with no Certificate of Innocence included (5). If the registered owner still contends they were not driving the vehicle, they must appear in court to contest the ticket. In 2007 the Portland Police Bureau started an electronic citation program. All the data required by the courts is now electronically sent to them which helps to increase the efficiency of both the courts and the photo radar program (4). Less staff time is now required. Public Education and Information Prior to the start of the program in 1996, Portland conducted an extensive public education campaign to inform and educate the public about photo radar as a tool for speed enforcement. Part of the campaign included outreach to the media through press conferences, radio, news- papers, and cable access televisions. They also provided informative material in the form of newsletters and direct mailings to reach local residents. PBOT manages a traffic safety and liability hotline. Concerns that are called in by the public are relayed to the Police Bureau, who then decides an appropriate action to take. The hotline receives more than 600 requests for speed enforcement a year (6). The Police Bureau also does education missions. As part of the requirements set by the legislature for the program’s trial period, the City of Portland conducted a public opinion poll in September 1996. The poll showed 74% approval by residents of using photo radar in neighborhoods, 89% approval for use in school zones, and 88% awareness that photo radar is used as a police speed enforcement tool (1). During the trial period PBOT set up a photo radar hotline to yield comments and concerns from the public. The

Case Studies 37 hotline received 789 calls during the first nine months of the program, 58% of which were calls expressing support for the program. Another study in 2003, conducted by a private firm, showed 87% of Portland residents were concerned about speeding. In 2005, the same firm conducted a telephone survey of 400 Portland residents. Sixty-eight percent of respondents agreed with the use of photo radar in school zones, and 85% responded that they would drive slower all the time if they saw photo radar being used at least three times per week. Program Evaluation Starting in 2005, the Oregon Revised Statute that authorizes photo radar requires cities to report once each biennium to the Legislation on the effects of the photo radar system on traffic safety, the degree of public acceptance, and the process of administration (4). According to the 2007–2008 biennium report for Portland, the number of speed violations monitored decreased by 5.3% from 2007 to 2008, while the number of enforcement hours increased. Table H-1 shows the program numbers from 2007 to 2008. Three of the top five deployment locations in 2008 were posted school zones, while the other two locations had a history of speed-related crashes and speeding complaints. Officer Frolov, a full-time photo radar enforcement officer with the Portland Police Bureau, observed a residual effect throughout the city since the start of the program. He noted that at the start of the program in 1996 there would be 250 speeding violations at a single location; now, at the same location, there are only 60 speeding violations (2). In 2005 an independent evaluation study was conducted by Christopher M. Monsere from Portland State University using data from the photo radar vans. The data was from 1996 through 2004 and included the total number of vehicles passing the van, the number of citations issued, and the percentage of vehicles passing the van that were in violation of the speed limit. During the time frame the number of vehicles passing the photo radar vans increased by approximately 6.5%. However, the number of speed limit violations decreased by 5.8%, and the number of issued citations dropped by 3.6%. Similar to the anecdotal observations of Officer Frolov, this study provided evidence of the historical impact that the program has had on speeds in Portland (7). Fiscal Considerations The automated enforcement program is costly to operate and generates a modest amount of excess revenue. The fee from the citations is split between the court, the county, and the city. Any excess revenue that is generated goes back into the program. The fee structure is established in the vendor contract and is part of the public record. The fee structure for the vendor has changed over time as the program evolved, with the issuance of new contracts. Item Year - 2007 Year - 2008 Enforcement Hours 2,602 2,713 Vehicles Monitored 1,118,811 1,208,048 Violations Captured 44,044 41,706 Citations Issued 27,018 22,904 Table H-1. Photo radar program numbers, 2007–2008 (4).

38 Automated Enforcement for Speeding and Red Light Running Red Light Camera Program In 2001, after 5 years of successfully using automated enforcement as a tool to reduce speed- ing, the City of Portland expanded their automated enforcement program to include red light running enforcement. Program Administration and Structure The Portland Red Light Camera (RLC) Program is operated jointly by the Portland Bureau of Transportation and the Portland Police Bureau. PBOT leads the site selection process, analyses, and most media relations, while the enforcement and review of citations is conducted by the police. Both organizations work closely with the vendor, ACS. As with the speed enforcement, the red light running program is a driver responsibility program. The violations are issued to the registered owner of the vehicle if a good quality photograph is captured and the gender of the observed driver matches the registered owner. Operations Portland currently has 11 red light cameras at 10 intersections throughout the city. The first six cameras were installed between October 2001 and April 2003 at five intersections. Five more cameras were installed between October 2007 and August 2009. There are no “dummy” cameras in the city, and the locations of the RLCs do not rotate between camera housings. Ten of the cameras use color film. These cameras take one picture of the vehicle in advance of the violation line and one picture of the vehicle beyond the violation line (in the intersection). In 2009 Portland installed its first digital RLC. The digital camera takes a front and rear picture of the violating vehicle, plus a 12-second video. Both the pictures and the video can be viewed online by the violator. Site Selection Candidate intersections for RLCs are selected by PBOT based on crash history, primarily related to crashes involving disregard of the traffic signal. Intersections undergo a formal evalu- ation performed by a private consulting firm to determine if the location is suitable for an RLC. The evaluation of each site follows Oregon Department of Transportations’ Red Light Running Camera Guidelines (8). Other considerations when selecting RLC sites include clearance inter- vals, offsets, signal timing plans (both peak and non-peak periods), spacing, conspicuity, and existing infrastructure (9). Warning Periods There is a 2-week warning phase with each new installation of an RLC. If a violation occurs during the timeframe at the new location, the registered owner is sent a warning in the mail. PBOT also writes a press release notifying the public of the new camera location (2). Citation Process The citation process for the RLCs is very similar to that for the photo radar. The cameras are oper- ated by a contractor, currently ACS. ACS processes the film from the cameras daily. They screen the photographs and retrieve owner data for the vehicles observed in the violations. ACS gathers the registration information for the violating vehicle from the Department of Motor Vehicles databases. The information is then posted to a secure Website for viewing by the Portland Police Bureau. A traffic officer trained in photo enforcement views the evidence and determines whether or not to issue a citation; this is done for each individual violation. If a citation is issued, it is mailed to the current registered owner within 10 working days of when the violation occurred.

Case Studies 39 Approximately 50% of the observed violations result in a citation being issued. For example, during 2008, 18,083 observed violations were processed, resulting in 8,767 issued citations. The biggest factors for non-issuance are gender match failure (19% of total citations not issued), and no front plate (16%), and clarity of driver (9%) (10). The State of Oregon’s vehicle code for adherence to the yellow interval can be classified as a restrictive yellow state. That is, a vehicle must stop on yellow unless it is unsafe to do so (ORS 811.265). However, the Police Bureau’s policy is more lenient. A citation is not issued unless the vehicle entered the intersection on red. Vendor Role As with the photo radar program, ACS is responsible for the daily processing of the photos and citations, as well as the maintenance of the camera equipment. ACS develops, digitizes and reviews the film, collects registration information, conducts the quality assurance/quality con- trol, and mails the citations. They also operate a call center to respond to questions and com- ments from the public. The vendor works directly with both the Portland Police Bureau and the Portland Bureau of Transportation and is in contact with both organizations on a daily basis. Adjudication If the registered owner receives the citation in the mail and he or she was not the driver at the time of the violation, the registered owner has the option to fill out and return a Certificate of Innocence, including a photo copy of their driver’s license. If the photo on the driver’s license does not match the photo from the RLC, the citation is dismissed. If the registered owner of the violating vehicle is a business (i.e., not a private citizen), an Affi- davit of Non-Liability is sent to the owner. The business can then either pay the violation fine or identify the driver at the time of the violation. If the business identifies the driver, a citation is re-issued and mailed to that driver (10). Public Education and Information The City of Portland ensures that all traffic signs at the equipped intersections conform with Oregon law and Manual on Uniform Traffic Control Devices (MUTCD) standards. Each enforced approach has a traffic signal ahead warning sign (W3-3) with a supplemental plaque that reads, “PHOTO ENFORCED.” An example approach sign is shown in Figure H-6. In addition to the signs at each equipped intersection, each major entering route into the City of Portland has a “TRAFFIC LAWS PHOTO ENFORCED” sign (R10-18), as illustrated in Figure H-7. Program Evaluation Requirements The City of Portland is required to provide a biennial report to the legislature on the program. There is no systematic follow-up for any of the sites; however, PBOT does do monitoring on an as-needed basis. Results The most recent biennial report was submitted in 2009. At the time of the report, there were ten cameras in operation at nine intersections total. The report provided the results of a before and after evaluation of the impact of the cameras on violation rates and crashes. The crash analysis

40 Automated Enforcement for Speeding and Red Light Running Figure H-6. Advance warning sign for red light camera enforced intersection. Figure H-7. Advance warning signs of photo enforcement upon entering city limits.

Case Studies 41 considered crash type and severity. The evaluation used the most recent 4 years of data before the cameras were installed, and 4 years of data after they were installed for eight of the nine inter- sections. For one of the intersections, only 3 years of before-and-after period data were available. The evaluation found a reduction in red light violations at all equipped intersections. The amount of reduction varied by intersection between 69% and 93% reduction in violations. The report concludes that Portland’s experience has been positive and that very positive trends are occurring at equipped intersections, pointing to reduction in injury crashes and reduc- tions in red light violation crashes. Fiscal Considerations As with the photo radar program, the payment structure for the RLC program is established in the contract. In the most recent contract period, the payment structure changed. Originally, the city paid the cost of the RLC installation. Now the vendor pays for the cost. However, the locations are still determined by the Police Bureau and PBOT. The vendor payment structure is a blended contract. The vendor receives a fixed amount per intersection and an amount based on the number of citations that are issued. The marginal amount decreases with more citations. The current payment structure is $27 per citation for the first 500 paid citations in a month, $20 for citations 501-700, and $18 for each paid citation over 700 in a month. Lessons Learned Improving Issuance The vendor, ACS, tracks the factors resulting from non-issuance of citations. The distribution of these factors is illuminating. Table H-2 presents the distribution for 2008. As reported in the biennial report, the issuance rate of citations increased from 41% in the first year of operation (2002) to 53% in 2004. By tracking the factors affecting non-issuance, the Police Bureau was able to improve their enforcement operations and establish better procedures. This in turn increased their issuance rate. Driver Anticipating Signal One problem noted by the Police Bureau was drivers anticipated the green signal and there- fore triggered the camera. The digital cameras, with better photo quality and video capability, should help eliminate some of the guess work in determining if the vehicles were in the inter- Factor # Citations Not Issued % of Total Not Issued No Plate 1504 16% Gender Match Failure 1736 19% Clarity of Driver 811 9% Framing of Car 537 6% Issuance Criteria Not Met 437 5% DMV No Hit 379 4% Emergency Vehicle 483 5% Glare on Windshield 611 7% Dark Interior 661 7% Clarity of Plate 197 2% Other 1960 21% Total 9316 100% Table H-2. Factors resulting in non-issuance of citations (10).

42 Automated Enforcement for Speeding and Red Light Running section during the red light. Agencies should be aware of this potential problem and be ready on how to handle this issue with the public. Accessibility to the News Media Periodically, the Portland program has been scrutinized by the news media or publicly scru- tinized by citizens receiving violations. The City of Portland has maintained tremendous acces- sibility to the public and to the news media. If a problem with the program is alleged, the city responds immediately with a press conference or a similar open response. Cautious Evolution Portland’s program, although operating for 15 years, is just now evolving to digital photog- raphy. Similarly, Portland has very cautiously and prudently expanded the program, opting in both cases to favor a slower evolution. This has helped them to maintain the public trust and avoid problems that other agencies have faced (e.g., public scrutiny of selecting sites for revenue generation). Recommendations for Starting a Program During interviews with key personnel of Portland’s photo radar and red light camera pro- grams, we asked what recommendations they would give to agencies looking to start an auto- mated enforcement program. The following lists recommendations and advice from the vendor, the Portland Police Bureau, and the Portland Bureau of Transportation. Vincent Parke, Northwest Regional Program Manager, ACS: • Identify a need—the program needs to be about safety; you need to have an existing problem or you will not have violations and you will end up losing money. • Need support of the Police Department. • Determine the best solution for your agency. • Communication with the client is key. • Get the right stakeholders involved. Peter Koonce, Signals and Street Lighting Manager, Portland Bureau of Transportation: • It’s all about trust with the public. • Make sure you have your objectives right—do you have a safety problem? • Get people with experience to support you and do a study (not as credible if the city does their own study); need an independent party and a second set of eyes to check your work. • Get constituents on board with the program. • Be mindful of what your revenue looks like. Do not set up the fee structure so you get lots of money—it will look bad to the public. Sgt. Todd Davis, Portland Police Bureau: • Pick a good vendor. • Be consistent. • Be receptive to public input. Robert Burchfield, City Traffic Engineer, Portland Bureau of Transportation: • Make sure you don’t abuse the public trust. • Safety focus—be honest about the safety; it is not about traps or fooling people; be conserva- tive with the number of cameras you have. • Identify stakeholders (community, police, and first responders) and tailor the program so it meets their interests.

Case Studies 43 References 1. Photo Radar Demonstration Project Evaluation, Cities of Beaverton & Portland. Portland, Portland Office of Transportation, 1997. 2. Davis, Sgt. Todd and Officer Frolov. Personal interview. Portland, August 19, 2010. 3. Burchfield, Robert M. Personal interview. Portland, August 19, 2010. 4. Davis, Sgt. Todd. City of Portland Photo Radar Project Report 2007–2008. Portland: Portland Police Bureau, 2009. 5. Parke, Vincent. Personal interview. Portland, August 18, 2010. 6. Raisman, Greg. Personal interview. Portland, August 19, 2010. 7. Monsere, Christopher M., Robert L. Bertini, Ahmed M. El-Geneidy, Tom Moes. Exploring Spatial and Temporal Performance Measurement in Metropolitan Transportation Safety Improvement Programs. London: Paper pre- sented at the 9th International Conference on Computers in Urban Planning and Urban Management, 2005. 8. Kittelson & Associates, Inc. Memorandum to Portland for Intersection Evaluation for Potential RLC Instal- lation. Portland: Kittelson & Associates, Inc., 2007. 9. Koonce, Peter. Personal interview. Portland, August 19, 2010. 10. Burchfield, Robert M. City of Portland Red Light Running Camera Program Biennial Report. Portland: Portland Bureau of Transportation, 2009. City of Virginia Beach, Virginia, Automated Enforcement Program Overview The City of Virginia Beach, Virginia, started its first automated enforcement program in Sep- tember 2004 with red light running cameras. Nine months later, the Virginia General Assembly allowed the automated enforcement legislation to end by sunset clause and the program was discontinued. The legislation was enabled again in 2007, and by March 2009, Virginia Beach became the first jurisdiction in the state to restart their program. The success of Virginia Beach’s red light camera program is apparent, as their program’s structure, operation, and public rela- tions are being replicated in cities and towns across the state. The following document presents an overview of Virginia Beach’s program and illuminates some of the noteworthy practices present that could be replicated by other agencies in the develop- ment and operation of an automated enforcement program. This information is based on inter- views with key personnel associated with Virginia Beach’s program and published documents. Background Enabling Legislation In 1995 the Virginia State Legislature approved the use of red light running cameras for a period of time that would end by a sunset clause in 2005. In 2005, the Virginia General Assembly voted not to renew the use of the cameras. As a result, Virginia Beach, as well as other localities across the state, had to discontinue the use of their red light camera systems (1). In 2007, the Virginia State Legislature voted to reauthorize the use of red light camera enforce- ment under State Code 15.2-968.1. The new legislation allows localities to install and operate red light cameras at one intersection for every 10,000 residents. With the new legislation came require- ments for the implementation and operation of the red light camera programs (2). Some of the key state regulations on photo enforcement included, but are not limited to • No monetary penalty imposed shall exceed $50, nor shall it include court costs. • If a locality does not execute a summons for a violation of this section within 10 business days, all information collected pertaining to that suspected violation shall be purged within 2 business days.

44 Automated Enforcement for Speeding and Red Light Running • No locality shall enter into an agreement for compensation based on the number of violations or monetary penalties imposed. • Before the implementation of a traffic light signal violation monitoring system at an inter- section, the locality shall complete an engineering safety analysis that addresses signal timing and other location-specific safety features. • All traffic light signal violation monitoring systems shall provide a minimum 0.5 second grace period between the time the signal turns red and the time the first violation is recorded. • Any locality that uses a traffic light signal violation monitoring system shall evaluate the system on a monthly basis to ensure all cameras and traffic signals are functioning properly. • Prior to, or coincident with, the implementation or expansion of a traffic light signal violation monitoring system, a locality shall conduct a public awareness program (2). One of the biggest differences between the 1995 legislation and the 2007 legislation is the involvement of the Virginia Department of Transportation (VDOT). Under the new legislation localities need VDOT approval for each red light camera installation. VDOT’s involvement in the site selection process for each automated enforcement program in the state helps ensure the cameras are being installed for the right reasons, which in turn helps boost public approval. Problem Identification Red light running was a growing concern in Virginia Beach. The City Council was first approached about starting an automated enforcement program in 1997, but it was not until 2004 that the first red light camera was installed. Red light running violation decreased dur- ing the 10 months of that initial program, but after they had to stop the program in 2005 due to state laws, violations began to rise again. Since the beginning, the goal of Virginia Beach’s program has been to change driver behavior. Changing driver behavior to reduce, and hopefully eliminate, red light running will ultimately make for safer roads and intersections. After the state legislation re-enabled the use of auto- mated enforcement in 2007, Virginia Beach began the process of restarting their program. The underlying problem and reason for restarting their program was red light running violations contributing to a high number of crashes. The police department did not have enough officers to conduct the continuous amount of enforcement that would be needed to achieve a change in driver behavior. With a high number of right turn on red (RTOR) violations, they expanded their original program to include enforcement of right turn lanes. Initiation of Enforcement The Traffic Engineering Division first approached the City Council in 1997 about starting an automated enforcement program; the program was tabled at that time. The City Council was approached again in February 2002, this time by the Police Department’s Special Operations Bureau. In May of the following year, the City Council approved the start of a red light camera program and allocated funds to the program. Mike Shahsiah, Senior Traffic Engineer for Virginia Beach, suggested that to increase success of the program being supported, initiation should come from the police department rather than traffic engineering (3). The initial program, which ran from September 2004 through June 2005, enforced a total of 30 lanes at eight different approaches. This program did not enforce RTOR violations. The current program, which has been in place since March 2009, enforces 106 lanes at 13 different intersections, including 36 left, 54 through, and 16 right turn lanes. It is important to note that the legislation allows the City of Virginia Beach to have automated enforcement at 43 intersections based on the population of the city (i.e., one intersection for

Case Studies 45 every 10,000 residents). However, the city only currently uses it at 13 intersections because these are the intersections that have been identified as having an intersection safety program where automated enforcement can help to reduce violations. Red Light Camera Program Program Administration and Structure The Virginia Beach Police Department leads the red light camera program. They work closely with the Virginia Beach Traffic Engineering Department for site selection and intersection analyses. VDOT also has a stake in the program, as it is required by state law to have VDOT approve each site before installing any automated enforcement. In reality, there are many other stakeholders and people involved with the program (see Figure H-8). This graphic illustrates the complexity of the program structure and the importance of collaboration between all the different stakeholders. Virginia Beach’s program operates as an owner liability program, as mandated through the state legislation. Violations are issued to the owner of the vehicle. The penalty for a red light violation is a $50 fine. This is civil penalty, meaning there are no driver license points assessed and no insurance implications. The Police Department serves as the lead agency for Virginia Beach’s red light camera pro- gram. The police run the program, which includes the day-to-day operations, public relations, maintaining the program’s Website, and media relations. They also work with Traffic Engineer- ing in the site selection process. Virginia Beach currently has six officers with responsibilities dedicated to the red light program, as shown in Table H-3. In addition to the staff in Table H-3, they also utilize the Police Department’s public relations and marketing officer to deal with all media correspondence. The vendor, Redflex Traffic Systems, provides the equipment for the red light camera system. Unlike some other programs, the vendor does not have a large role in the operations or manage- ment of this program. In discussions with the Virginia Beach Police Department, they believe that as the end user, the Police Department should be the lead agency, and the vendor should have little involvement (4). The vendor is simply providing a service. The selection of sites, pub- Figure H-8. Stakeholders involved in a red light camera program (Source: Officer Brian Walters).

46 Automated Enforcement for Speeding and Red Light Running lic information, structure, and general operation of the program should be driven by the Police Department in coordination with traffic engineering and other partners. The Police Department should have control of the program. Operations Virginia Beach currently has red light cameras monitoring 20 approaches at 13 intersections. The cameras monitor all lanes on an approach—left, through, and right turn lanes—for a total of 106 lanes. All cameras were activated between March and December 2009. The digital cameras capture three rear photographs of the violating vehicle—one prior to the stop bar, one beyond the stop bar, and one close-up picture of the rear license plate. A video camera also records a 12-second video that captures 6 seconds before the violation and 6 seconds after the violation. The City of Virginia Beach has established that their cameras will only be activated if a vehicle crosses the loop sensors at a designated speed of 15 mph or above after the light has turned red. The city does not publicize this tolerance. The vendor had recommended a lower speed of 12 mph, which would result in more citations. However, the Virginia Beach program staff were concerned that this did not reflect the program’s goal of improved safety and that all decisions related to the operation of the program should reflect that goal. They also found that moving the loop sensors closer to the stop bar reduces the number of false triggers. There is a half second grace period after the light turns red (Virginia Beach refers to this as the amnesty period). The system is set up so that it does not activate during the period of 0.00 to 0.49 seconds after the light turns red. All violations occurring during this time are not captured. The State of Virginia currently mandates the highest grace period in the country. However, this means that there are vehicles entering the intersection up to 0.49 seconds into the red that do not receive any form of citation or warning. The Police Department still uses traditional enforcement of signal violations at the intersections. In one comparison situation, the automated enforce- ment system identified one signal violation for every ten identified by traditional enforcement due entirely to the amnesty period. Advance warning signs, as shown in Figure H-9, are posted within 500 ft. of the enforced intersections on all approaches, even if all the approaches are not monitored. Gateway signs (Figure H-10) are also posted on the major thoroughfare entering the city to warn drivers that Virginia Beach uses automated enforcements. Current Staffing Job Description 1 Sergeant: Program Supervisor Supervises Red Light Camera Program, Alarm Reduction & Telephone Reporting Unit. 1 Master Police Officer: Program Coordinator Public Awareness, Presentations & Media Spokesperson; Program Assessment & Reports; Intersection Safety Analysis; VDOT Approval and Annual Recertification Packages; Coordinates program with REDFLEX, Traffic Engineering & Traffic Operations; completes administrative correspondence; process violations (see description below). 1 Master Police Officer: Program Manager 3 Part-Time Officers Review (accept/reject) violations; process vehicle license plates/tags; mail correspondence – affidavits, hearing requests, undeliverable, etc.; interact with citizens by phone, walk-ins, etc.; contact owners for court, prepare court packages, and attend court; process unpopulated violations; utilize Accurint to locate current or valid addresses and readdress to resend notice. Table H-3. Virginia Beach Police Department red light camera program staffing.

Case Studies 47 As previously noted, the systems use loop sensors in the pavement. Working with the Traffic Engineering Department, the Police Department explored the use of other sensor technologies. They tried flush mount sensors and light detection and ranging (LIDAR). Neither provided the desired accuracy. Site Selection As part of the Virginia State Code, site selection should be based on four factors: 1. Crash rate for the intersection. 2. Rate of red light violations at the intersection. 3. Difficulty of law-enforcement officers to patrol the site and apprehend violators. 4. Difficulty of law-enforcement officers to apprehend violators safely and within a reasonable distance. The Police Department and Traffic Engineering work together to identify high crash locations at signalized intersections and locations with a high number of red light running citations. These two lists are combined to find the sites that overlap. Traffic engineering then reviews each crash report for these sites, identifying crashes by approach. Required by state code is a safety analysis of the intersection, which includes addressing the signal timing and other location-specific safety features (2). Virginia Beach takes the review of the potential intersection a step further and initially conducts an informal road safety audit (RSA) of each location. The Virginia Beach Police Department organizes the RSA team, con- sisting of enforcement, engineering, design, and other specialists as needed, depending on the unique characteristics of the intersection. They may also involve local business owners. The RSA team considers the safety performance of the intersection from the perspective of all road users, including unfamiliar drivers, older drivers, pedestrians and cyclists. The Virginia Beach Police Figure H-9. Advance warning sign for red light camera enforced intersection. Figure H-10. Advance warning sign of photo enforcement upon entering city limits (Source: Officer Brian Walters).

48 Automated Enforcement for Speeding and Red Light Running Department has found these RSAs to be an incredibly useful tool to improve the locations of candidate intersections. The RSAs provide a low-cost assessment of prospective intersections and also help to increase the collaboration of all the involved parties and public relations. After the detailed safety analysis is completed, an application packet is sent to the VDOT Dis- trict Engineer for review. Virginia State Code requires all intersections with photo enforcement to be approved by VDOT. Following the review, the city and the District Engineer discuss any changes that need to be made, and the application is then sent to VDOT headquarters for final approval. During the review, VDOT takes a rigorous look at the whole intersection, not just the approaches with proposed cameras. Both Traffic Engineering and the Police Department agree that the VDOT review of every intersection is an advantage to the program because it adds another layer of review, and it insulates the city from any potential lawsuits (3–4). Warning Periods There is a 30-day warning period at all sites after the cameras are installed. If a violation occurs during the warning period, a warning notice is mailed to the violator. The warning period is not only a way to educate the public about the cameras, but is also helps find any glitches in the sys- tem before citations are issued. It also helps send a message to the public that the program is for safety reasons, not monetary reasons. This warning period is also used to work with the media to educate them about the program. The initiation of a system at an intersection generates a lot of curiosity and interest from the media in covering the story. Each new installation is a public education opportunity and is used to encourage changes in overall driver behavior beyond the enforcement intersections. During the warning period the program coordinator and traffic engineering staff are continu- ously monitoring the reported violations, looking for any irregularities and comparing them to field observations. For example, there was one location where there was an abnormally high number of right turn on red violations during the warning period. After observing this location in the field, they noted the stop bar was functionally set back too far from the intersection and crosswalk. Vehicles were coming to a stop beyond the stop bar. The advance loop detectors were too far back so the cameras were triggered and registering these as violations. The team relocated the right turn lane stop bar closer to the intersection, shown in Figure H-11, resulting in accurate enforcement. Without the warning period, this discovery would have taken place after multiple violations would have been issued and challenged in court, creating unnecessary burden on many departments. Figure H-11. Location where right turn lane stop bar was relocated after observing a high number of false violations during the warning period.

Case Studies 49 Citation Process Citations are issued by police officers viewing pictures and video collected through the auto- mated process. Since all the cameras are digital, the process of putting together the picture and video packages is done automatically by a computer. When the packages are complete, a sworn police officer reviews each violation. Each photo includes a data bar which displays the speed, yellow time, time into red, and the date and time of the violation. Most of the quality assurance and quality control comes from the officers checking the data bar. If a signal timing issue is found then the intersection is flagged and the violations are rejected. The benefit of the doubt always goes to the violator. The intersection is also immediately reviewed to resolve the problem. Fifty-eight percent of total violations captured by the cameras are thrown out. The biggest reason for dismissal is the vehicle completed a safe turn on red. Table H-4 shows the top five reasons for dismissing a violation. After the officer approves a violation, the violators are sent a citation package in the mail. This includes the three photos taken of their vehicle during the vio- lation, as well as a link to the Website where they can view the 12-second video of the violation. The citation must be issued within 10 business days from the time of the violation. If not, state legislature mandates that all information collected pertaining to the suspected violation must be purged within 2 business days (2). Adjudication By state law, Virginia Beach operates as an owner liability program. Violations are a civil pen- alty and can be challenged in the civil court. Virginia Beach is required by the State Code to provide instructions in the citation on how to file an affidavit if the owner of the violating vehicle wishes to declare his or her innocence. Initially, violators were brought to traffic court, but after receiving complaints from lawyers, the proceedings were moved to civil court by the city. The use of video technology substantially assists with the adjudication process. Several judges have voiced their appreciation of the video; it provides a strong base of evidence and allows the judges to view the violation in its entirety. The judges are actively involved in the red light camera program and interested in the camera effects on driver behavior. Discussions with the judges resulted in the installation of additional warning signs specifically designed toward reducing right turn on red violations. The Virginia Beach Police Department believes it is very important that agencies brief their judges on the system activation and process being used in capturing violations in order for the judges to have confidence in the system’s fairness and increase their commitment to uphold the issued violations (4). Public Education and Information As part of public education and outreach to the community, Virginia Beach developed their own name and logo for the program: PHOTOSafe Virginia Beach. Referring to the program as Table H-4. Top reason for violation dismissal from January 1, 2010, to December 31, 2010. Reason for Dismissal Percent of Total Captured Violations Safe turn on red 30.32 % License plate obstructions 5.44% Police discretion 3.30% Emergency vehicles 2.58% Vehicle obstruction 2.02%

50 Automated Enforcement for Speeding and Red Light Running PHOTOSafe helps emphasize to the public that the program is about safety, not about increasing revenue for the city, a common misconception. Virginia Beach uses the PHOTOSafe branding, which includes the logo shown in Figure H-12 and the slogan “Red means stop!” in every aspect of public information and outreach. Both the logo and slogan are incorporated into the advance warning signs noting the use of photo enforcement when vehicles enter city limits. A majority of the public education is done through the use of the program’s Website: www. vbgov.com/photosafe. The Webpage provides information about the PHOTOSafe program, including where the cameras are located, how they operate, frequently asked questions, signal timing information, city and state ordinances and codes, as well as links to related information. Officer James Barnes, Jr., the public relations and marketing officer, said they often get Freedom of Information Act (FOIA) requests for information on their program. If enough FOIA requests come in for the same information, they will often make the information available on their Website for all to see and use (5). The police department makes a strong effort to put as much information as they can on the Website because it helps reduce the number of phone calls they receive. However, they also aim to make the Webpage as low maintenance as possible since they do not have a dedicated staff member to make frequent updates. Information on the Website is updated as needed, but the objective was to create a Website that could stand on its own with minimal maintenance so that resources can be directed towards other aspects of their program. In addition to the Website, the police department makes frequent presentations about their program to civic organizations and various groups, as well as making program information readily available to the media. Transparency and accessibility is important to the success of the program and general public acceptance. Program Evaluation Frequency of Evaluations Once a year, Virginia Beach is required to re-certify each intersection with VDOT, as are all agencies in Virginia with programs. This involves re-checking the signal timings and noting any changes that may have occurred to the geometry. The state also mandates a monthly evaluation of each camera system. The purpose of this evaluation is to ensure that the signals and cameras are functioning properly. Whenever an intersection signal is taken offline (e.g., for modifications or special events), traffic engineering staff download the signal data overnight to make sure the signal was Figure H-12. Virginia Beach PHOTOSafe logo and slogan (Source: Officer Brian Walters).

Case Studies 51 put back online correctly. Incorrect signal timing would affect camera operations, so staff are continually monitoring this to ensure there is not any improper function. Staff monitor not only the signal timings but also the rate of violations. Sharp increases in violations on a given day indicate there may be a problem with the system set-up and/or the signal timings. Any erroneous signal timings would result in the dismissal of all violations at that location for that time period. This check has helped to identify and resolve problems in an immediate and efficient manner. Results There has not been a full safety analysis of this second installation of Virginia Beach’s red light camera program since it has only been operational since the spring of 2009. However, the Police Department and Traffic Engineering have been tracking the number of violations at each intersection. One location, shown in Figure H-13, has seen a 69% reduction in violations over the first 16 weeks after cameras were installed. The first safety assessment of the program will be in 2011. This will involve a comparison of 1 year of crash data before camera installation (2008), 1 year during camera installation (2009), and 1 year after camera installation (2010). A more in-depth analysis will be available in 2013, comparing 3 years of before-and-after crash data. The results of these studies will determine if the city needs to expand the program to more intersections. That is, the city prudently will use the results of the crash analysis to determine the effect on crashes and use that to inform their decisions on expansion or any modifications to the program. The police officers and traffic engineers acquire and compare violation rates for every 1,000 vehicles traveling in each lane that is enforced. For example, a detailed analysis of the data of the Great Neck Road approach shows that the right turn lane had reductions in violations from a rate of 8.0 (Sept. 2009), to 5.5 (May 2010), to 4.0 (Aug. 2010), as shown in Table H-5. The violation rate indicates that the number of violations occurring in the right turn lane was reduced by 50% in 1 year. In other words, statistical data indicates that for every 1,000 vehicles Figure H-13. Sixteen-week comparison of red light running violations at Virginia Beach Blvd. and Great Neck Road (Source: Officer Brian Walters).

52 Automated Enforcement for Speeding and Red Light Running traveling in the right turn lane, four vehicles are probable to make the turn on red without stopping. Figure H-14 displays comparable violation processing information over a 13-month period. Month one on the figure is September 2009, the month that the camera was activated. Warning letters were sent to drivers during this month. The change in the number of violations cap- tured, to number of violations mailed to registered owners, represents the number of violations reviewed and rejected by police officers. The data displayed in this table does not include the number of violations occurring during the 0.5-second amnesty period. The Traffic Engineering Department conducts ongoing assessments of the PHOTOSafe pro- gram by collecting and reviewing crash data at the enforced locations as part of their annual intersection crash assessment report. They are also working with the psychology department at Old Dominion University in the completion of a scientific study to determine the effects the photo enforcement cameras are having on crashes. Violation Rate = Number of Violations / Traffic Volume ë 1,000 Dates of Sample Number of Violations Traffic Volume Violation Rate 6/1/09 – 9/2/09 7,659 957,082 8.0024 9/30/09 – 5/4/10 10,909 1,968,916 5.5406 5/5/10 – 8/4/10 3,618 891,021 4.0606 Table H-5. Right turn lane violation reduction calculations at Great Neck Road. N um be r o f V io la tio ns 2674 2050 1947 0 500 1000 1500 2000 2500 3000 1 1848 1366 1158 2 1399 1219 1030 856 886 793 3 4 1434 11 1 1159 742 1008 699 5 6 Months of Operation 66.37% reduction in the number of Violations Captured 67.26% reduction in total Available for Prosecution 65.84% reduction in number of Notices Printed and Mailed 4 1289 824 779 7 1102 974 792 692 754 666 8 9 1026 1073 723 773 712 746 10 11 1053 782 775 12 899 671 663 13 Figure H-14. Thirteen-month comparison of red light running violation data of the Great Neck Road approach at the intersection with Virginia Beach Blvd. (Source: Officer Brian Walters).

Case Studies 53 Evaluation of Right Turn on Red Traffic Control Devices As previously noted, one of the reasons Virginia Beach started the automated enforcement program was the high number of red light violations for right-turning vehicles. The police department, traffic engineering staff, and even the court judges have collaborated on how to address this problem and modify driver behavior. One of the solutions was to try different warning signs to emphasize the right turn on red enforcement. The current signage, shown in Figure H-15, is a “photo enforced” placard underneath the “Stop here on red” sign and is located next to the stop bar. Previous efforts, shown in Figure H-16, included the “Stop here on red” Figure H-15. Current photo enforced warning sign for right turns. Figure H-16. Previous sign efforts to enforce right turn on red.

54 Automated Enforcement for Speeding and Red Light Running sign without the “photo enforced” placard, and a “Right on red after stop” sign with “photo enforced” in red letters underneath. The police have already noticed a higher compliance rate at certain locations with these signs. These locations tend to be where there is less visual clutter around the intersection. They are using data from the cameras, including the data collected dur- ing the amnesty period, to determine the most appropriate sign to use that influences drivers to stop on red. This effort is a great example of how to use not only the cameras to change driver behavior, but how to use the data the cameras provide to change other traffic control devices to influence driver behavior. Fiscal Considerations The program generates modest revenue from the citations. In the first 18 months the pro- gram collected $2.45 million. As part of the vendor payment agreement, $1.31 million was sent to the vendor. The remaining $1.14 million went into the city’s general fund and can be used to fund traffic safety improvements. A well-run program is expensive to maintain. Currently, the Virginia Beach contract cameras cost $4,350 per month for monitoring up to five lanes and $4,740 per month for monitoring up to seven lanes. The number of lanes affects not only the time and effort needed, but also the cost to operate. The cost to pay for the system is covered by three to five violations per day. The payment recovery rate from citations is 68%. The program operation costs are paid out of the Police Department budget. One hundred percent of the money collected is deposited in the city’s general fund. As such, the budget for the program and the revenue it generates are independent funding streams. The agency staff notes that this separation is necessary in order to maintain the integrity of the program, clearly associating the program with safety. Virginia Beach’s program uses two full-time police officers. One of the officers is the pro- gram coordinator who interacts with the public, media, VDOT, camera vendor, and city traffic engineering and operations departments. The second is the program manager who runs the violation processing components of the program. This ensures that the program is under tight control and that the operation of the program is closely monitored in the interest of the travel ing public. Additionally, three retired officers are employed to review the citations. The retirees are sworn officers in the department with duties strictly limited by the Chief of Police to photo enforcement. This reduces the program costs and releases officer resources for other patrol duties. Other Notable Practices Training Sessions for Other Agencies The Virginia Beach program has generated a lot of interest from other municipalities in Virginia that are exploring the use of automated enforcement as a tool in their jurisdictions. Virginia Beach receives frequent requests for information about their program. This prompted the program manager to develop a course entitled PHOTOSafe 101. The 1-day course was pre- sented in April and May of 2009, and again in 2010, for agencies interested in learning about the structure of their program and the process of proposing an intersection for automated enforcement to VDOT. The course included the following topics: • Overview of Virginia Beach’s program with capture and processing demonstrations. • Review of Virginia’s photo enforcement law and the importance of public awareness. • City and VDOT expectations and the public’s perception of photo enforcement. • Review of VDOT guidelines and the completion of Engineering Safety Analyses. • Discussion on Request for Proposal (RFP) selection processes and varied technology used in photo enforcement.

Case Studies 55 Virginia Beach has been overwhelmed by the attendance at the course. Agency representa- tives drove to Virginia Beach from all over the state to gather information on the program. This included representatives from police departments, traffic engineering departments, and indi- viduals at VDOT interested in learning more about the program. The demand for the course illustrated the lack of information available about structuring a program and conducting inter- section reviews. Based on the demand for this course, it was given a second time. Virginia Beach continues to provide this information when requested. Opportunity for Observation The automated enforcement system provides an opportunity for observation of driver behaviors in response to the roadway environment. As previously discussed, the City of Virginia Beach has used their systems to make observations about the best signing to convey RTOR prohibitions. The system allows them the ability to evaluate the impact of different signs on driver behavior. They have also observed that downstream merges increase violations in the outside lanes, and that intersections with less visual clutter have better compliance. Regional Consistency As noted in the previous section, other jurisdictions in the state have reached out to Virginia Beach for information on their program structure. This includes the neighboring cities of Newport News and Chesapeake. Virginia Beach provided these agencies their program materials including their program branding. The two agencies now have a similar look to their signing, logos, and pro- gram materials as Virginia Beach. The media in the Hampton Roads area of Virginia refer to the red light camera enforcement as the “PHOTOSafe program.” It appears to drivers that the PHOTOSafe program is a regional program versus an agency-specific program. This likely helps to increase any spillover or halo effect as it appears to broaden the scope of each individual program. In addition to sharing a similar look, the three programs also have a similar structure and process. This is critical to the Virginia Beach program manager. Before allowing these agen- cies to use Virginia Beach’s materials, he met with each of them to ensure that their program would follow the same stringent structure. Consistency is necessary to preserve the integ- rity of the program, as any scrutiny of the program in any other jurisdiction would also be directed at Virginia Beach. Drivers’ seeing the PHOTOSafe as a regional program strengthens its credibility. Public Information Officers in the region meet every other month to discuss common issues, including the PHOTOSafe programs. The collective group is called the Hampton Roads Media Council. Participating jurisdictions provide accessibility to the public through the media and use the media as a tool for educating and information sharing. This practice of regional consistency benefits both the agencies and the driving public. For the driving public, drivers know what to expect as far as policies and practices regarding automated enforcement. They also see standardized signing. For the agencies, the program development costs are greatly reduced by using the pioneer agency’s (in this case, Virginia Beach) materials. Additionally, learning from the pioneer agency’s existing practices helps the agencies to have a successful, well-structured program operating within months. Unfortunately, in some regions, neighboring agencies may be interested in having programs for the potential revenue. If so, agencies should consider the implications of regional consistency. Employing Officer Walter’s practice of meeting with the program managers and reviewing their structure may help to avoid potential problems. Problem Identification and Resolution When problems are identified at intersections, the program coordinator uses the same multi- disciplinary process that was employed in selecting and reviewing the initial intersections to

56 Automated Enforcement for Speeding and Red Light Running resolve the issue quickly and in the best interest of the traveling public. An example of this occurred when a review of the systems identified that occasionally a camera view was blocked. An investigation revealed that when the air temperature was high, a cable at the intersection would expand and hang into the view of the camera. The program coordinator initiated a multi- disciplinary field review to explore the problem and identify a resolution. The field review team included representatives from the police department, traffic engineering, the vendor, and the cable company. By having the various agencies in the field at the same time, they were able to collectively explore different solutions together, greatly reducing the time from problem iden- tification to problem solution. They considered raising and lowering the camera, raising and lowering the traffic signals, moving the camera to another pole, and moving the cable to another pole. For each proposed solution, the viability of the proposal, the impact on the intersection safety and operation, and the cost were considered. The resolution of the problem was to install a near-side traffic signal. Lessons Learned The Virginia Beach program coordinator and program partners have thought through and explored their program in great detail. Their attention to detail and the structure of their pro- gram has helped them to identify many lessons learned for other agencies. These include the following notable points for other agencies. • Continue Other Interventions. RTOR violations can be a pervasive problem at some inter- sections and the movement may represent the greatest number of violations. The signing at the intersection should clearly convey to the driver that they are required to stop on red. The city has tried multiple sign variations to communicate this message. The stan- dard signing in the MUTCD does not appear to be adequate. The city is tracking drivers’ responses to different signs to identify the optimal signing configuration to reduce the number of these violations. The automated enforcement data also provides a great method to measure drivers’ responses to the different signs. When the optimal sign is identified, it can be used at other intersections as well. Although this is still a work in progress, the important lesson here is that the city continues to use a comprehensive approach that includes signing and markings to reduce violations at their intersections. They do not simply install automated enforcement and allow the violations to continue without attempt- ing other interventions. • Heavy Vehicles. Most intersections in the program capture rear images. However, for locations with high truck traffic (10% or more is suggested by Virginia Beach staff as a good threshold) front cameras are used to capture the license plate on the truck unit rather than the trailer. The license plate on the trailer differs from the front license plate. The front plate is the plate that corresponds with the truck’s owner. • Communication. “Positive PR is of the upmost importance.”—Officer James Barnes, Jr. Good communication with the media and the public, especially during the beginning of the program, saves a lot of time and helps to dispel concerns or misunderstandings about the program. Providing detailed information on the Website (e.g., signal timings at each inter- section in the program) or through other readily available program material reduces the number of calls and requests for information. One person should be identified for all media requests to maintain a consistent message and to establish a relationship with the media. He or she should meet with the media and agency public relations early in the process. Ask the media to meet you at the sites so that you can explain the system structure and illustrate the safety problem that the program is trying to address. This will likely have a two-fold benefit, in that the media will be better educated to report on your program and their cover- age of your program will help to educate the public about the dangers of signal violations at all intersections.

Case Studies 57 • Comprehensive Approach. Automated enforcement is only one part of the package to improve intersection safety and will contribute a portion towards safety improvement at the inter section. Road Safety Audits, good traffic engineering, appropriate design, traditional enforcement, and public education should all be employed to improve intersection safety. Additionally, the courts and judges should be involved in the process. Program staff should proactively provide infor- mation to the court system regarding the program and should be available to respond as needed to any questions that may arise. • Continue Traditional Enforcement. Automated enforcement does not take the place of an offi- cer; the systems are there to enforce when the officer cannot be there. Officers should continue to provide traditional enforcement at intersections with red light cameras. Recommendations for Starting a Program During interviews with key personnel of Virginia Beach’s red light camera program, we asked what recommendations they would give to agencies looking to start a similar program. The fol- lowing lists recommendations and advice from the Virginia Beach Police Department and the Virginia Beach Traffic Engineering staff. • The public must have the following three items: – Public knowledge of the systems (how do they work?). – Public awareness of the systems (where are they located?). – Public assurance of the systems (will it stop red light running?). “There are three keys to success: public knowledge, awareness, and assurance.”—Officer Brian Walters • The intersection reviews are critical. The yellow and red timings have to be right. • Make sure you do all your homework. Make sure you are selecting the appropriate locations. • Coordination with traffic engineering, traffic operations, and the police department is critical. “You need teamwork to make it happen.”—Mike Shahsiah. • The program initiation should come from the police department, not traffic engineering. The program will have more support if it is initiated as an enforcement function, not a traffic engineering function. • The number of lanes that are enforced has a large impact on the time, effort, and cost to operate the program. When planning your program, do not think purely of the number of intersections or approaches that will be enforced, consider the number of lanes. • Be prepared for FOIA requests and media requests. Share these requests with the police department and other agency staff. • Assign people roles within the program (e.g., media relations and technical aspects). • If possible, establish uniformity between neighboring jurisdictions if multiple programs exist in a region. References 1. PhotoSafe Virginia Beach FAQs. PhotoSafe Virginia Beach. [Online] Accessed 23 August 2011, http://www. vbgov.com/photosafe 2. Virginia State Code 15.2-968.1. Use of Photo-Monitoring Systems to Enforce Traffic Light Signals, Richmond, VA: Virginia General Assembly, 2007. 3. Shahsiah, Mike. Personal interview. Virginia Beach, October 27, 2010. 4. Walters, Officer Brian. Personal interview. Virginia Beach, October 27, 2010. 5. Barnes, Jr., Officer James. Personal interview. Virginia Beach, October 27, 2010. 6. Virginia Department of Transportation, Red Light Running Camera (Photo Enforcement) Engineering Safety Analysis Guidelines, Traffic Engineering Division, Richmond, VA, 2009.

58 Automated Enforcement for Speeding and Red Light Running City of San Diego, California, Red Light Photo Enforcement Program Overview The City of San Diego, California, began operating the current red light photo enforce- ment program in 2002. They operate 15 red light running cameras in the city. The program is a joint effort between the San Diego City Council, Traffic Engineering Division, and Police Department. The program was designed and implemented using solid principles, as well as lessons learned from the city’s initial photo enforcement program. This document will provide an overview of the program as it exists today. Best practices will be identified that could benefit other agencies as they develop automated enforcement programs. The information contained in this report is based on published reports and personal interviews with key personnel involved in the operation of the program. Background The Safe Lights for San Diego program is very different from the automated red light camera program that was first operated in San Diego in 1998. The current program is a testament to how well a red light camera program can be implemented in a community that has significant reservations about automated enforcement. In 1996, the State of California enacted California Vehicle Code Section 21455.5, which autho- rized local government entities to use automated photo enforcement systems at intersections. In 1998, San Diego initiated a Red Light Photo Safety (RLPS) program. The program was met with public concerns. A class action suit was filed against the program. On September 4, 2001, Ronald Styn, Judge of the Superior Court of California, affirmed a lower court decision that evidence from the red light camera program could not be admitted as evidence in court. The decision was broadly based on the red light camera contractor having too much control over the program combined with a financial incentive to issue more citations. The judge determined that the vendor was actually operating the program, not the City of San Diego as required by law. The court considered the fact that the vendor had moved inductive loops used for incident detection without the knowledge of the city. The vendor also conducted the installation, cali- bration, and maintenance of the camera equipment. The city did not inspect the camera system even after construction was completed. The judge determined that this combination of events tainted the evidence. The judge also ruled that because the vendor was essentially operating the program and being paid on a contingency basis there was a potential conflict that undermined the trustworthiness of the evidence. Based on this court ruling, approximately 250 red light camera citations in San Diego were dismissed (1). Since that time, the City of San Diego and the State of California each completed red light camera program audits. The California State Auditor Report stated that, “although they have contributed to a reduction in accidents, operational weaknesses exist at the local level,” and made specific recommendations for San Diego and other cities to improve their programs. The report encouraged more rigorous oversight of vendor operations (1). The California audit went on to question if the San Diego camera locations were chosen with safety as the top criteria. They reported 5 of the 19 red light camera sites were not on the most dangerous intersection list. The report recommended that other engineering improvements be considered before a red light camera was installed at any specific intersection. Also, San Diego had not provided the vendor with specific written business rules about when to issue a citation, relying on verbal communication only (1).

Case Studies 59 Positive news coming out of the California audit was the reduced crash experience related to the red light camera program. San Diego reported an 8% decrease city wide in red light running crashes after the implementation of the program. At intersections in San Diego equipped with red light running cameras, crashes decreased by 16% (1). It is reasonable to conclude that the San Diego community would be predisposed to be skep- tical of a new red light camera program. This report will focus on the structure of the new red light camera program. If a red light camera program can be successfully launched in San Diego after the experience of the initial program, a red light camera program has the potential to be successfully implemented in many other communities across the United States. Problem Identification A published list of the most dangerous intersections in San Diego provides the list of candidate intersections with high red light running crash experiences. Other candidate sites, in addition to those on that list, are selected based on high red light running crashes (2). Enabling Legislation California Vehicle Code Section 21455.5 was enacted in 1996. This law holds the driver of the vehicle responsible for a red light violation documented by a camera program. The law require- ments include the following key components: 1. Identification of the system by signs visible to traffic approaching from all directions, or posted signs at all major entrances to the city. 2. An intersection with a system must have a minimum yellow light interval established by the Department of Transportation. 3. Must make a public announcement of the program at least 30 days prior to commencement of enforcement. 4. May only issue warning notices for the first 30 days of program operations. 5. Only a government agency, in cooperation with a law enforcement agency, may “operate” an automated enforcement system. An automated enforcement operation includes: a) Developing uniform guidelines for screening and issuing violations, processing and stor- ing confidential information, and establishing procedures to ensure compliance with those guidelines. b) Perform administrative and day-to-day functions that include: i. Establishing guidelines for site selection. ii. Ensure the equipment is inspected regularly. iii. Certify the equipment is properly installed, calibrated, and operating properly. iv. Regularly inspect and maintain the warning signs. v. Oversee the establishment of and any change to signal phases and timing. vi. Maintain controls necessary to assure that only those citations that have been reviewed and approved by law enforcement are delivered to violators. vii. Items i., iv., v., and vi. in the above list may not be contracted out by the governmental agency but item iii. may be contracted out to an equipment supplier or provider if the governmental agency maintains overall control and supervision of the system. 6. Photographic records made by an automated enforcement system shall be confidential and shall be made available to governmental agencies and law enforcement agencies only for the purpose of this law. 7. A contract between a governmental agency and an equipment supplier or manufacturer may not include provision for the payment or compensation based on the number of citations generated, or as a percentage of the revenue generated as a result of the use of the equipment authorized.

60 Automated Enforcement for Speeding and Red Light Running Safe Lights for San Diego The current red light camera program was launched in November 2002, with a new 5-year agreement with ACS. The new program was designed around the specific recommendations and lessons learned from the San Diego and California audits. Many of the San Diego politi- cians did not support starting a new red light camera program. Mr. Jon Hannasch, an engineer in San Diego’s Traffic Engineering Division, manages the program for the city. According to Mr. Hannasch, the City Manager supported the new program, and after many public meetings the city gained approval to start a new program by one vote (2). Program Administration and Structure The program relies on a close working relationship between the San Diego Traffic Engineer- ing Division and the San Diego Police Department. The Traffic Engineering Division is the lead agency, although most of the work is done by the San Diego Police Department. Mr. Hannasch of the Traffic Engineering Division described this configuration as an effort to gain more con- trol over the program (2). Before an automated enforcement program would be initiated again in San Diego, the City Manager wanted to ensure he had control over the program. The Traffic Engineering Division reports to the City Manager but the San Diego Police Department does not. While many parts of the San Diego Government play a role in the program, these two agen- cies are the most involved. Operations San Diego currently operates red light cameras at 15 intersections. Each of these camera loca- tions are identified on the San Diego Red Light Photo Safety Program Website (3). The system uses sensors buried in the road surface to identify vehicles that enter the intersection against a red signal. The system automatically photographs the front and rear of each vehicle with close up images designed to identify the driver and registration plate of the vehicle. In addition, the system also captures a 10-second video of the vehicle entering the intersection. Figure H-17 shows a diagram of the red light camera system. Figure H-18 is a close-up image of an actual camera in the field. Site Selection The most dangerous intersections, based on collision history and collision type, are reviewed by the Traffic Engineering Division to determine potential sites for the red light cameras. If one of those intersections experiences a high number of red light running crashes it is placed at the top of the list of potential sites. Input is accepted from the San Diego Police Depart- ment, which keeps a running list for implementation at a future date. Traffic volume increases and red light violation numbers are also considered, but the decisions have all been made with red light running crashes as the top criteria. Current red light camera locations, as well as red light camera locations under consideration, are posted on the San Diego Website (3). This Website notification allows citizens to offer input about the potential future camera sites. There are over 1,500 signalized intersections in the city, but only 15 (1%) have red light camera systems. Mr. Hannasch said he was surprised that more of the most dangerous intersections in San Diego did not have higher red light running crash experience (2). By focusing on the intersections that are both on the most dangerous intersection list and have a high number of red light running crashes, Mr. Hannasch expects that the number of red light camera systems will remain a low percentage of the overall signals. He is currently exploring the possibility of moving the camera systems to different locations instead of adding systems (2).

Case Studies 61 Figure H-17. Red light camera system location (3). Figure H-18. San Diego red light camera system.

62 Automated Enforcement for Speeding and Red Light Running Warning Periods San Diego issued warnings for the first 30 days of the program. They also issue warnings for 30 days after each new red light camera installation. Warning periods are set by the California Vehicle Code Section 21455.5. Citation Process Images that are automatically captured at red light camera locations are electronically sent to Arizona where they are reviewed by vendor personnel. Following specific business rules estab- lished by the City of San Diego, the vendor personnel make an initial decision as to whether an image may capture a violation of the law. The vendor obtains the registered owner information for California-registered vehicles. San Diego police obtain the registered owner information for out-of-state registered vehicles. The vendor puts the potential violation images onto a Website in Arizona (4). San Diego Police Department personnel examine each set of photo images to determine if an image supports a conclusion that a violation of the law took place. For vehicles registered in California, a San Diego police officer will compare the driver image in the violation photo to the California DMV driver’s license photograph of the owner. If the comparison of the violation image with the DMV license image is inconclusive, they will check other databases that may have photographs of the vehicle owner. Sgt. Joel McMurrin manages the program for the San Diego Police Department. Sgt. McMurrin described the face image identification as being critical to the process. He will not issue a citation unless they can identify three distinct identifiable points of comparison between the DMV image and the violation face image. They look at the nose, mouth, unusual marks, etc. to try to make a match. They also consider the time difference between the dates each of the two images were taken (4). Figure H-19 illustrates the process. For vehicles registered out of California, the San Diego police officer does not have a DMV image to compare with the violation image. Instead, the officer looks at the violation face image and issues the citation only if the gender and race match and if one could identify the driver in court based on the violation image. Sgt. McMurrin advised, “We would rather miss issuing ten citations than issue any notice of liability in error” (4). A San Diego police officer reviews both rear and front images to determine if all of the estab- lished violation criteria are met. They will review the 10-second video of the incident if there is Figure H-19. Sgt. Joel McMurrin reviewing a violation image with Officer John Labo.

Case Studies 63 any question about the violation taking place. According to Sgt. McMurrin, they want to decide to issue the citation based on the totality of the circumstances. Traffic Engineering Role The Traffic Engineering Division is responsible for the overall management and accountabil- ity of the program. They are responsible for making the site selection, requesting approval of any new location by the San Diego City Council, and the program operational policies. Before the grace period of 0.5 seconds into the red phase was eliminated in 2002, Mr. Hannasch had to describe the rationale behind the change. The ultimate decision to change the grace period was made by the City Council. They maintain and update the Website that describes the program to the public. They led the request for proposal process that ultimately selected American Traffic Solutions (ATS) to replace ACS in 2008. They are responsible for managing the vendor contracts with input from the other city stakeholders (2). The Traffic Engineering Division checks for the proper placement of the warning signs and con- ducts system checks every month with the Police Department and the vendor (see Figure H-20). Police Department Role The San Diego Police Department is responsible for the issuance of notices of liability. Only a San Diego police employee may approve a notice of liability for issuance. For the review and approval of notices of liability, the San Diego Police Department uses police offi- cers that are assigned to the Traffic Division. At times, these are motorcycle officers, officers recovering from an injury, as well as parking enforcement officers. They never use civilian employees for notice of liability approvals. Figure H-20. Jon Hannasch conducts a system check on a red light camera system.

64 Automated Enforcement for Speeding and Red Light Running The officers that testify in court for red light camera cases are always motorcycle officers that conduct traffic enforcement operations of all types on a regular basis. They progressively receive more training on the red light camera operations. Training includes an annual day-long train- ing session, with monthly updates from the vendor on the camera system operation, as well as teachings on the legal aspects related to the red light camera program. The Police Department conducts monthly system inspections with the vendor and the Traffic Engineering Division. They check for the proper placement of the warning signs and conduct system checks. Vendor Role San Diego currently works with ATS. ATS is responsible for installing and maintaining the red light camera systems at the locations selected by the San Diego Traffic Engineering Division. They must inspect every red light camera system a minimum of twice a month. ATS is responsible for the automatic capture of violation images at each red light camera sys- tem. They transmit each image to the ATS processing center at the ATS headquarters in Arizona. At the center, ATS personnel review each set of images based on the business rules established by the City of San Diego. They access DMV records for vehicles registered in California and several other states and then incorporate the appropriate data in a record attached to the vehicle images. If the images meet the established business rules, they post the combined violation record to their Website to wait in a queue for a review and approval decision by a member of the San Diego Police Department. ATS personnel print and mail notices to appear that have been approved by the San Diego Police Department. They prepare evidence logs in support of adjudication cases when citizens request a court trial. They manage citizen inquiries for issues such as lost notices to appear, how to request a court date, and how to pay a fine. Vehicle Owner Role The owner of a vehicle who receives a notice to appear has a few decisions to make. He or she can either accept responsibility and pay the fine or file a form stating his or her innocence. The minimum fine is currently $480 for a violation. The owner has up to 45 days to request a hearing on the alleged offense. If the owner was not the driver of the vehicle at the time of the offense, he or she could return a form at least 10 days prior to the court appearance date that describes the issue and identifies the actual driver. The San Diego police would then be able to issue a notice to appear to the actual driver of the vehicle. Adjudication A San Diego police officer testifies in court for red light camera citations if a citizen requests a trial. According to Deputy City Attorney, Melissa Ables, this is the same practice that is used in San Diego for other traffic cases (5). Sgt. McMurrin advised that more individuals are scheduled for trials at the same time than the court can manage. Officers triage the cases and meet with defendants before the case is called for trial. Officers consider each case and decide if they would have issued a traffic citation themselves had they witnessed the violation in person. Officers are permitted to amend the violation to a lesser offense with a lower fine or dismiss the case at this point. If a defense attorney requests information in advance of a scheduled trial, and an officer is anticipating a complicated defense, they can request that a representative from the City Attorney’s Office be available for court.

Case Studies 65 Recent Court Issue On August 16, 2010, Karen A. Riley, Commissioner of the San Diego Superior Court, issued a decision that dismissed eight red light camera cases. The main issue in the motion to dismiss the cases related to the evidence prepared by ATS for the initial court cases. It included hearsay and 6th Amendment Confrontation Clause objections (6). The court determined the data on the data bar superimposed on each red light violation image was an accurate representation of the data in the computer, but there was no presumption that the data itself was accurate or reliable. The court concluded that if contested, the data on the vio- lation images could not be admitted into evidence unless a person could offer foundational evi- dence that the computer was operating properly. The judge decided since there was no evidence presented to support a finding that the computer itself was operating properly, the information imprinted on the photographs would be excluded in the eight cases (7). The judge also determined that evidence logs prepared by ATS for the initial court trials were inadmissible as evidence. The judge determined that they were not actual logs prepared con- temporaneously with the act or event; they were more like reports dated several months after importing the images. They appeared to have been created solely for the purpose of litigation. She decided that they did not meet the foundational elements for either hearsay exception (7). In conclusion, the court decided that several parts of the ATS affidavit, the evidence logs, and the superimposed data on the violation images, were not admissible without a live witness in court to testify to their knowledge of the contents. The judge dismissed the cases concluding the people would be unable to prove the eight cases beyond a reasonable doubt without this evidence (7). Sgt. McMurrin advised that, since this negative decision, a revised training program be estab- lished. The San Diego Police Department began annual training put on by ATS and monthly update training with all officers who process and testify in court on photo red light cases. Since this training has been conducted, very few cases are ever lost at trial. Those that are found not guilty are usually based on identity issues. Even experienced traffic court defense attorneys do not wish to go to trial due to the professionalism and knowledge of the San Diego Police officers who testify in court on these cases. Most trials now are by individuals who just want to plead their case in front of a judge and take their chances. A more recent court decision found that everything in the evidence package submitted by ATS to include the Field Service and Inspection logs were admissible as evidence. All images, video and documentation, were found not to be hearsay (4). Similar cases are still being debated in other locations that operate automated traffic enforce- ment programs based on criminal laws. The Superior Court of Orange County, California, dis- missed seven red light camera cases in an August 19, 2010 decision (8). A Superior Court of the State of California, County of San Bernardino, filed a similar decision on December 21, 2010 (9). These issues are likely to be raised in other jurisdictions in the United States that operate automated traffic law enforcement programs. The original precedent for the recent cases was the United States Supreme Court case Melendez-Diaz v. Massachusetts (10). The case was argued before the Court on November 10, 2008, and decided on June 25, 2009. The defendant in the case had been arrested in possession of a white powder. A lab analysis report was submitted as evidence that described the white powder to be cocaine. The Court decided that the affidavit submitted by the lab analyst was not admissible as evidence unless the analyst was available to be questioned about the testing methods that he had utilized. Public Information and Education San Diego wants their community to know they are operating a red light camera program. Prior to the new program being implemented in 2002, the city held many public briefings and

66 Automated Enforcement for Speeding and Red Light Running meetings. Major changes to the program continue to be discussed in open hearings, and the media continues to publish stories about the program. The San Diego Website describes the program operation, documents where the systems are installed, and answers frequently asked questions. Signs are posted at each red light camera equipped intersections (see Figures H-21 and H-22). Fiscal Considerations The San Diego red light camera program provides a flat fee to ATS for their services. This eliminates the concern about a vendor having an incentive to issue more notices of liability, plac- ing all of the financial risk on the City of San Diego. As a result, the Traffic Engineering Division must ensure the contract with ATS has strong language in reference to required performance measures. ATS is responsible to pay liquidated damages if their system is not functioning as required by the contract. The Traffic Engineering Division has the burden and responsibility to carefully manage the performance of ATS. Mr. Hannasch reported that the San Diego red light camera program had recently passed the breakeven point, with revenues surpassing the program costs. San Diego receives $152.68 from every paid Notice of Liability. This money is used to pay for the program operation, including payments to ATS and city personnel salaries. The rest of the money goes to the State of California and to the courts. The increases in the overall fine amount have been the result of additional fees, security costs, and other fees not directly associated with the program. Figure H-21. Photo enforced signs are posted at every red light camera location. This one faces drivers on the approach to the intersection being enforced.

Case Studies 67 Lessons Learned • The San Diego experience has demonstrated that a successful red light camera program may be implemented in a location that has already had a negative experience with a red light camera program. • The program must be operated with the objective to improve traffic safety. • Media outreach and an open public hearing process is important to ensure the community understands the safety focus of the program. • Paying a vendor a flat fee is important for eliminating concerns about compromised evidence, but it necessitates strong contractual performance requirements. Program Legal Suggestions to Consider In a roundtable discussion with Mr. Hannasch, Sgt. McCurrin, and Deputy City Attorney Ables, the following items were suggested as potential future changes to consider for the California law: • An owner liability law would be easier to manage and could reduce the burden on the courts. • The law could be changed to allow a criminal offense for flagrant first violations (based on the time into red) and all subsequent violations, with civil violations for other first offenses. • Current California law for the misuse of a disabled person sticker allows an officer to decide if it should be charged as a misdemeanor or a ticket based on the violation circumstances; this could be used as a model for a potential new red light camera law. Figure H-22. This photo enforced sign faces drivers approaching the intersection from the side that is not enforced.

68 Automated Enforcement for Speeding and Red Light Running • The current law only permits the violation images to be used for the purpose of providing a red light running violation. This should be adjusted to allow for the images to be used during the investigation and prosecution of major crimes, serious crashes, missing persons and similar cases. Currently the law prohibits the use of these images even if the information would be exculpatory. Recommendations for Starting a Program • Start with problem identification. • The program must focus on safety. • Red light camera systems should be designed to maximize crash reductions. • Ensure the agency, not the vendor, is in control of the program. • The vendor should be compensated based on a flat fee. • Operate a program as transparent to the public as possible. • Communicate to the public the safety effectiveness of the program. • Personnel testifying in court should be well trained. References 1. California State Auditor, “Red Light Camera Programs: Although They Have Contributed to a Reduction in Accidents, Operational Weaknesses Exist at the Local Level,” Bureau of State Audits, Sacramento, CA, 2002. 2. Hannasch, Jon. Personal interview. San Diego, CA, October 11, 2010. 3. “Red Light Photo Safety Program,” City of San Diego. Website accessed March 5, 2011, http://www.sandiego. gov/engineering-cip/services/public/rlrphoto/ 4. McMurrin, Sgt. Joel. Personal interview. San Diego, California, October 11, 2010. 5. Ables, Melissa D., Deputy City Attorney. Personal interview. San Diego, CA, October 11, 2010. 6. Cella, Matthew, “Court Rulings Give ‘Go Slow’ to Red-Light Cameras,” Washington Times, Vol. 30, August, 2010. 7. Riley, Karen A., “8 City of San Diego Photo Red Light Cases, Motions to Exclude Evidence Packets,” B16464A, B16681A, B16772A, B17833A, B17968A, B17983A, B18095A, B18404A, s.l., Superior Court of California, County of San Diego, August 16, 2010. 8. Prickett, Gregg, G. Lewis, and K. Robinson, People of the State of California v Khaled, 30-2009-304893, Superior Court of California, Santa Ana, CA, May 21, 2010. 9. Cohn, David, J. Brsico, and W. J. Powell. People of the State of California v Macias. ACRAS 900155, s.l., County of San Bernardino Superior Court, December 21, 2010. 10. Supreme Court, Melendez-Diaz v Massachusetts (No. 07-591). Cornell University Law School. Accessed online March 6, 2011, http://www.law.cornell.edu/supct/html/07-591.ZS.html City of Edmonton, Alberta, Canada, Automated Enforcement Program Overview The City of Edmonton in Alberta, Canada, has multiple automated enforcement programs including covert and overt mobile photo radar, conventional red light cameras, and inter- section safety cameras which form part of a speed management continuum. The initial Red Light Camera program started as a pilot project in 1998 at a single location. From 1999 on, Edmonton was adding 12 more red light camera sites every year until in 2003 the final num- ber of 60 sites and 24 rotational camera units was reached. Now, with new legislation allow- ing police to use intersection safety cameras to capture both red light running and speeding vehicles, the program covers 50 locations with 50 stationary Intersection Safety cameras, while conventional red light cameras are being phased out. The success of Edmonton’s program is apparent, with a recent Empirical Bayes evaluation study showing that the red light running

Case Studies 69 program has been effective in reducing the overall number of collisions and collision severity at most camera locations. The following document presents an overview of Edmonton’s program and illuminates some of their noteworthy practices that could be replicated by other agencies in the development and operation of an automated enforcement program. The information is based on interviews with key personnel associated with Edmonton’s programs and published reports. Background Problem Identification The City of Edmonton has identified 12 corridors that receive daily photo radar automated enforcement as part of an Integrated Corridor Safety Program. The selection of these corridors is based on collision problems identified by the City of Edmonton Office of Traffic Safety. Com- plementary to this enforcement program, two corridors are reviewed annually for engineering safety improvements. These reviews have been part of the program since 2009. Specific atten- tion is given to at least two identified high collision contributors, namely right turn cut-offs, and left turns across the path of oncoming vehicles. As the traffic safety reviews are done, identified changes are entered into a master engineering traffic safety priority list. This list includes a review of corridors against city-wide infrastructure to identify engineering changes that would provide a good cost-benefit ratio if changed, or can be accommodated into future capital budgets or rehabilitation programs. The initial foray into the red light automated enforcement program was piloted by the Edmonton Police Service (EPS) to help reduce the high prevalence of intersection-related collisions. Two aspects relate to this decision. First, over half of the collisions in Edmonton were intersection related; these contributed a disproportionate number of fatalities and seri- ous injuries which, at the time of the program introduction, was very high in Edmonton and across Canada. Second, the increased danger associated with manned enforcement of red light running necessitated the use of this new technology for officer safety reasons, as well as public safety (i.e., other drivers lawfully traveling through the intersection). In addition, manned enforcement augmented by mobile photo radar enforcement increases the impact of the deter- rence theory, which enhances driver perception of increased enforcement. Enabling Legislation In 2004 the City of Edmonton established a Mayor’s Task Force on Traffic Safety. The task force recommended the creation of a municipal Office of Traffic Safety (OTS). By late 2006, the OTS was established and created an Edmonton Traffic Safety Strategy that was reflective of the provincial and national strategies, but specific to Edmonton. One specific target of the Edmonton strategy was addressing the issue of speeding. In 2007, the Edmonton Police Service (EPS) requested that the City Council assume responsibility for the administrative aspects of the automated enforcement program from a private contractor. The City Council approved this request and the OTS became the lead agency for the administration of automated enforcement equip- ment, which at the time related to red light cameras and photo radar. The EPS retained respon- sibility for the direction of enforcement. In January 2009, provincial legislation was passed to allow police to use Intersection Safety Cameras (ISCs). ISCs capture both red light running and speeding vehicles. The guidelines state that ISCs detecting speed can only be used at intersections where there is also detection of red light running (1). In other words, enforcement of red light running and speeding is a package deal with the ISCs.

70 Automated Enforcement for Speeding and Red Light Running Current legislation does not allow for unmanned photo radar enforcement equipment. The Edmonton Police Service stated they would like to see a change to this legislation, so it may pro- vide them with the opportunity to install unmanned speed enforcement devices at mid-block locations and highways in addition to intersections. Initiation of Enforcement The Edmonton Police Service started using mobile speed enforcement in 1995 with a single vehicle. By 2008 the city had five photo radar vehicles. The first red light camera was installed in 1998. Eleven more red light cameras were added in 1999. By 2003 Edmonton had 24 cameras that rotated randomly between 60 sites. Photo Radar Program Program Administration and Structure The intersection safety camera and photo radar program is run jointly by the Office of Traffic Safety and the Police Service. At the request of the EPS, the administration of the program was re-assigned to the City of Edmonton. The OTS is responsible for the management and installa- tion of the cameras, as well as the engineering analyses for program evaluation. The EPS main- tains all enforcement-related responsibilities. Each organization has a lead role that is complementary to the process. This was specifically done to address the perception of automated enforcement as being purely a revenue generator or “cash cow.” This joint effort provides scientific basis for site selection by OTS, which is reviewed and approved by EPS from an enforcement context (2). Operations Edmonton has both covert and overt mobile photo radar vans. Currently there are ten covert vans, with three more being added this year. The new equipment allows for instantaneous feed- back to the operator of the van. The operators are able to see the images of the speeding vehicle on the laptops in the van and ascertain if the camera settings are correct. The ability to make field adjustments to the cameras during changing environmental conditions ought to improve image quality. It also allows the operator to immediately confirm their visual observations on the laptop. As data is electronically uploaded at the end of each working shift, the equipment also allows for more rapid processing of violations, thereby ensuring violators get their violations in a timelier manner. The city also has four additional vans, known as the Safe Speed Community Vans. These vans are designed to stand out to drivers, with photos of Edmonton citizens cover- ing the sides of the vans. They are located in areas that communities have identified as hot spots for speeding and are equipped with the same photo radar equipment as the covert vans. These vehicles are primarily used for educating drivers to be aware of their speed, but are also used to generate tickets for those who fail to comply with the speed limit. The Edmonton Police Service has established a speed tolerance for the photo radar, which var- ies depending on the circumstances of the location it is being utilized in. The tolerance used for the equipment is consistent with the ISC system and is the same tolerance given by most police agencies when manually enforcing speed. The photo radar speed enforcement is specifically targeted at 12 designated corridors for daily enforcement for a minimum period of time. Typically, seven to eight photo radar vehicles are deployed every day for two shifts. Photo enforcement signs are posted throughout the city based on the provincial guidelines.

Case Studies 71 Site Selection For both photo radar and ISCs, Edmonton currently uses several state-of-the-art Empirical Bayes based methods to prioritize sites. These methods include expected collisions, expected excess collisions, expected net collision reduction (using crash reduction factors for different crash types), and applying weights for collision costs. For photo radar, an additional method is used that estimates expected collision reduction based on anticipated reduction in mean speed, a procedure documented in NCHRP Report 617 (3). Edmonton does not yet know if one method is more successful than the others; they continue to look at sites highly ranked by any method, giving particular attention to sites that are highly ranked by more than one method. Warning Periods Alberta Provincial guidelines mandate a 4-week warning period for any automated enforce- ment program. During the warning period a notice is sent out to all violators notifying them of their infractions. The notice includes the violation data such as location, date and time of the offence, speed of the violating vehicle, fine amount, registered owner and vehicle information, and two images of the violation along with the image of the plate close up. If a new site is added during the initial warning period, a notice is sent out to violators at that location. Any new site that is installed after the warning period does not require warning notices to be sent out. Citation Process At the end of each deployment of the photo radar vans, the violations are uploaded and the images and data are synchronized. The data is sent to the vendor, ACS, for initial screening and license plate viewing. Following the screening, the data is returned to the EPS and then sent to Service Alberta for information on the registered owner of the violating vehicles. Service Alberta is a private-sector company that serves as a registry agent, providing motor vehicle services to the Province of Alberta. The data from Service Alberta is returned to the EPS, who then sends it to ACS. ACS conducts a blind verification of license plates. If the license plate data matches, then the violation is forwarded to the EPS in a violation review queue. The violations in the queue are then approved by a Peace Officer. Upon approval, the violation is printed and mailed by ACS. Vendor Contract and Payment Vendors were identified through a competitive bidding process. A request for proposal was issued and based on responses to the solicitation, a vendor was identified and a contract was awarded. In order to choose the best equipment, Edmonton looked at numerous factors such as compliance with all technical requirements, ease of use and minimal maintenance, developed training manuals and schedules, existence of customers with similar programs, good references, warranty period, and financial stability of the vendor. The equipment supplier is paid according to the equipment pricing list included in the contract. It is a one-time payment for each order based on a purchase order. Spare parts needed for equipment maintenance are also paid based on the contract pricing list and specific purchase orders. Back-end processing for photo radar and intersection safety camera tickets is paid on a monthly basis as a flat monthly fee plus additional fee per paid ticket. Up until September 2009, the outside vendor was responsible for all aspects of supplying the necessary equipment and services to operate photo radar and intersection safety cameras. This included equipment supply, installation, and maintenance. All of these services were paid for through the flat-fee portion of the monthly payment. The outside vendor was also responsible for back-end processing of photo radar and red light camera tickets and monthly reporting.

72 Automated Enforcement for Speeding and Red Light Running These services were paid for partially through the flat-fee portion and fee per paid ticket. The city is progressing towards management of all aspects of back-end processing. The contract period is 5 years for the equipment vendor and month to month for the process- ing vendor. Collaboration Several city departments and sections work together. This includes the City of Edmonton transportation department, transportation operations, OTS analytical staff, and provincial automated enforcement audit personnel. The EPS and the OTS have a standing equipment management committee. Speed manage- ment is handled by an integrated committee involving the EPS, the OTS, and transportation engineers. A data committee is responsible for the integration of required data across depart- ments and between agencies. The initial project to integrate new automated enforcement equip- ment and processes for the City of Edmonton is managed through an inter-departmental steering committee. Operational issues identified outside the scope of these committees are handled at a management level through direct contact and meetings as required. The collaboration between departments and agencies has proven successful in Edmonton. Both political will and a commitment of the chief of police and the general manager of the trans- portation department have reduced barriers and encouraged strong integrated and joint traffic safety plans that increase collaboration. Public Education and Information The Edmonton Police Service maintains a Website with information on the photo radar program: http://www.edmontonpolice.ca/TrafficVehicles/PhotoRadar.aspx. The Website pro- vides background information on the program, information on how to pay traffic fines, a list of photo enforcement sites, and answers to photo radar frequently asked questions. Currently, there is a gap between public perception of support for automated enforcement and actual support. In 2010, most adult Edmontonians (78%) either agree, or strongly agree, with using automated enforcement. However, only a third (34%) of the adult population believes that automated enforcement has public support. Red light camera programs have higher levels of support than automated enforcement activi- ties that address speeding. While 83% of Edmontonians agree, or strongly agree, with using red light cameras to ticket drivers who run red lights, 75% supported the use of intersection safety cameras to ticket drivers who speed through intersections. Even fewer Edmontonians (73%) supported the use of photo radar to ticket drivers who are speeding. Perceived support for specific automated enforcement programs parallels actual support. There is more perceived support for red light cameras than there is for intersection safety cam- eras and photo radar. While 45% of adult Edmontonians believe that red light cameras have public support, fewer believe that the population supports automated enforcement that targets speeding (33% perceived support for intersection safety cameras that ticket drivers who speed, and 30% for photo radar). Edmonton is working to address the gap between perceived support and actual support through social norms marketing. Social norms marketing combines social marketing techniques (using marketing techniques to change behavior rather than sell products) with social norms theory (which posits that most people’s behavior is influenced by their perceptions of what is normal or typical). If Edmontonians learn that most of the population supports the use of

Case Studies 73 automated enforcement, then they are more likely to openly support the technology and the behaviors that it deters. The OTS is working with the EPS to plan how to use social norms mar- keting to address perceptions about automated enforcement. Program Evaluation Traffic and violation data are collected during photo radar vehicle deployment. Based on data analysis and site performance (e.g., a decrease or increase of the speed-related problems) the site becomes permanent, continues in use from time to time, or is abandoned. There has been no formal crash-based evaluation of the program, but this is in the future. The Office of Traffic Safety is responsible for monitoring the program. The program is also subject to external audit and monitoring by provincial solicitor general employees. A quarterly report is prepared and provided to the province for audit purposes. Fiscal Considerations Revenue from automated enforcement is used to support the cost of the program as well as support traffic safety initiatives. In addition, Edmonton City Council approved the expenditure of 1.5 million Canadian dollars to establish a permanent Urban Traffic Safety Research Chair at the University of Alberta, Department of Civil Engineering. The funding for the research chair is from automated enforcement revenue. Any additional funds generated above the cost of operations can be dedicated to a standing list of traffic safety projects and priorities that have been established based on cost-benefit reviews. No additional surplus revenues are anticipated based on this use. Intersection Safety Camera Program Program Administration and Structure The Office of Traffic Safety provides recommendations for intersection safety camera loca- tions based on safety performance analyses, which include an analysis of the number of colli- sions. In addition to the recommendation of sites for ISCs, the OTS also manages all equipment purchased for the program, maintenance and repair for the equipment, coordinating construc- tion of new ISC sites, and working with the vendor. Similarly to the photo radar program, the EPS maintains all enforcement related-responsibilities. Operations Edmonton currently has intersection safety cameras monitoring 50 approaches at 28 intersec- tions. Twenty-three ISCs were installed in 2009 as a result of the new legislation. Twenty-seven more cameras were installed in 2010. The cameras monitor approximately three percent of the city’s 928 signalized intersections. The intersections did not undergo any improvements (e.g., optimization of the cycle length or yellow intervals) prior to the use of automated enforcement. All cameras are permanently installed; they do not rotate between multiple sites. In 1998, the program started out using wet film in all their cameras. The switch to digital equipment was made in September 2009 with the advent of the use of ISCs. The optimal number of cameras is dictated by the number of locations at which the safety benefits will outweigh the costs. Edmonton is currently working on determining this optimal

74 Automated Enforcement for Speeding and Red Light Running number. Edmonton police believe the fixed and mobile speed enforcement methods are comple- mentary to each other. The zone of influence of the fixed installations is currently being studied with a view to avoid the use of mobile enforcement in that zone. The City of Edmonton has established a grace period of 0.2 seconds after the light turns red. Violations are not issued for cars entering the intersection between 0.0 and 0.199 seconds. This is consistent for all intersections throughout the city. The reason a grace period was established is because at time 0.0 seconds both the yellow and red light is visible on the image. Edmonton police also found it difficult to enforce as drivers would state that the light had not been red for the time stated. They had strong public reaction when there was no grace period, but have since found that the public thinks the 0.2 second grace period is fair. The Edmonton Police Service has also established a speed tolerance for the ISCs, which can vary depending on the circumstances of the location of deployment. The tolerance used for the equipment is consistent with the ISC system and is the same tolerance given by most police agen- cies when manually enforcing speed. The Province of Alberta mandates that permanent gateway signs be placed on primary roads entering municipalities that use automated enforcement to alert drivers of the use of automated enforcement technology. There is also the requirement of a speed limit sign, a photo enforcement sign, and a type of enforcement sign (i.e., red light and/or speed camera) on all approaches to an intersection where the ISCs are in place. Site Selection The site selection is a joint effort between the Office of Traffic Safety and the Edmonton Police Service. The OTS reviews the collision statistics and any other available data to compare loca- tions and prepare a list of proposed sites. OTS technicians then visit the site and perform the site inspection to determine the feasibility of installation. After the site visits, the EPS reviews the revised list, approves certain sites, and then prioritizes the locations for ISC installation. As previously discussed, Edmonton currently uses several state-of-the-art Empirical Bayes based methods to prioritize sites. These methods include expected collisions, expected excess collisions, expected net collision reduction (using crash reduction factors for different crash types), and applying weights for collision costs. As a result of this site selection process, half of the existing ISCs are located on the 12 corridors that are part of the Integrated Corridor Safety Program. Warning Periods Alberta Provincial guidelines mandate a 4-week warning period for automated enforcement. Edmonton’s warning period lasted 2 months due to technical problems at the beginning. The police also did not send out a sufficient number of notices in time, so the warning period was extended beyond the 4-week mandate. During the warning period, notice is sent out to all violators notifying them of their infrac- tions. The notice includes the violation data and images. If a new site is added during the initial warning period, a notice is sent out to violators at that location. Any new site that is installed after the warning period does not require warning notices to be sent out. Vendor Contract and Payment Vendors were identified through a competitive bidding process. A request for proposal was issued and, based on responses to the solicitation, a vendor was identified and a contract was awarded. The request for proposal included both intersection safety cameras and photo radar components. The same vendor was successful in the competition bid for both programs.

Case Studies 75 Since the same vendor is used for both photo radar and ISCs, the payment process is the same. The equipment supplier is paid according to the equipment pricing list included in the contract. It is a one-time payment based on a purchase order. Back-end processing is paid on a monthly basis as a flat monthly fee plus additional fee per paid ticket. Up until September 2009, the outside vendor was responsible for all aspects of supplying the necessary equipment and services to operate the cameras. This included equipment supply, installation, and maintenance. All these services were paid through the flat fee portion of the monthly payment. The contract period is 5 years for the equipment vendor and month to month for the processing vendor. Collaboration Collaboration among the city departments for the ISC programs is similar to that for the photo radar program. In addition, the various city departments work together with the vendor and contractor to run a smooth program. Adjudication The legislation enables the collection of unpaid fines for violations that are adjudicated in court by attaching the fines to renewal of annual license plate fees and driver’s licenses. A screening tool is under development to identify multiple-offence violators for closer scrutiny by manned law enforcement personnel. Public Education and Information Edmonton held a media conference before the initiation of the intersection safety cameras. The city received feedback that the media and public were very supportive of the cameras. Similar to the photo radar program, they created a Webpage on the city’s Website that contains information about the cameras, camera locations, and frequently asked questions: http://www.edmonton.ca/ transportation/roads_traffic/intersection-safety-cameras.aspx. The Website explains that Edmonton uses automated enforcement at high-collision intersections to discourage drivers from running a red light or speeding, to lower the number and severity of collisions, and to improve the safety of Edmonton’s roads. It also provides statistics on collisions at intersec- tions. The OTS provides evidence-based statistics for collisions in Edmonton. In addition to the city’s Website on ISCs, the Edmonton Police Service also maintains a Web- site: http://www.edmontonpolice.ca/TrafficVehicles/IntersectionSafetyCameras.aspx. These pages provide background information on the program, information on how to pay traffic fines, a list of camera locations, and answers to frequently asked questions. The OTS and EPS are also prepared to respond to media inquiries about the ISCs and the program. To raise awareness of the enforcement cameras, Edmonton organized a competition between the local high schools for the best drawing to be painted on the camera poles. As a result, 12 camera poles were painted by the winners. The camera poles that were painted were all in close proximity to schools so that students could be reminded every day about the role of the cameras and the consequences of poor driving behavior. As with photo radar, there is a similar gap between public perception of support for ISC’s and actual support. Less than 30% of the population believes that the ISC program has public support. It appears that over 86% may actually support the program. Red light camera pro- grams have higher levels of support than intersection safety cameras. Edmonton is working to address this through positive social marketing. A marketing campaign is targeted for develop- ment in 2011.

76 Automated Enforcement for Speeding and Red Light Running Program Evaluation The Office of Traffic Safety is responsible for monitoring the program. The program is also subject to external audit and monitoring by provincial solicitor general employees. A quarterly report is prepared and provided to the province for their audit purposes. Edmonton has not yet reached the formal evaluation stage for their current ISC program as it is relatively new and the deployment of sites is still in progress. A planned evaluation will also study the effects of the ISCs on surrounding intersections and corridors. The Edmonton Police Commission has completed a peer-reviewed study that has established the benefits of the original red light program. The study report is available to the public on their Website: www.edmontonpolicecommission.com. Results from this study show a reduction in the total number of collisions at most sites after the cameras were implemented, as well as a reduc- tion in the collision severity (4). Fiscal Considerations Revenue from the intersection safety camera program is distributed in the same manner as revenue from the photo radar program. All revenue from automated enforcement is used to support the cost of the program as well as support traffic safety initiatives. In addition, the revenue is also used to support the Urban Traffic Safety Research Chair at the University of Alberta. Any additional funds generated above the cost of operations can be dedicated to a standing list of traffic safety projects and priorities that have been established based on cost-benefit reviews. No additional surplus revenues are anticipated based on this use. Notable Practices Notable practices from Edmonton’s automated enforcement program include: • The unique cooperation arrangement between the EPS and the OTS. The OTS provides the scientific expertise for crash-based site selection and the EPS does the enforcement. • The use of statistical rigor in site selection for both ISC and mobile photo radar. They use crashes, Safety Performance Functions, target crashes, and crash costs in methodology based on the Highway Safety Manual (HSM) procedures. • The use of statistical rigor in program evaluation. The Empirical Bayes (EB) methodology has been used to evaluate both programs. • Coordination of the use of ISCs with mobile photo radar. A research project is underway to determine the influence zone for ISCs to ensure mobile photo radar locations will not overlap. References 1. Province of Alberta, Automated Traffic Enforcement Technology Guidelines, January 2009. 2. Edmonton Police Service. Personal interview. Edmonton, 2010. 3. Harkey, David L, et al., NCHRP Report 617: Accident Modification Factors for Traffic Engineering and ITS Im- provements, Transportation Research Board, Washington, D.C., 2008. 4. Sayed, Tarek and P. de Leur, Photo Enforcement Traffic Safety Study, Edmonton Police Commission, Edmonton, 2006.

Abbreviations and acronyms used without definitions in TRB publications: AAAE American Association of Airport Executives AASHO American Association of State Highway Officials AASHTO American Association of State Highway and Transportation Officials ACI–NA Airports Council International–North America ACRP Airport Cooperative Research Program ADA Americans with Disabilities Act APTA American Public Transportation Association ASCE American Society of Civil Engineers ASME American Society of Mechanical Engineers ASTM American Society for Testing and Materials ATA American Trucking Associations CTAA Community Transportation Association of America CTBSSP Commercial Truck and Bus Safety Synthesis Program DHS Department of Homeland Security DOE Department of Energy EPA Environmental Protection Agency FAA Federal Aviation Administration FHWA Federal Highway Administration FMCSA Federal Motor Carrier Safety Administration FRA Federal Railroad Administration FTA Federal Transit Administration HMCRP Hazardous Materials Cooperative Research Program IEEE Institute of Electrical and Electronics Engineers ISTEA Intermodal Surface Transportation Efficiency Act of 1991 ITE Institute of Transportation Engineers NASA National Aeronautics and Space Administration NASAO National Association of State Aviation Officials NCFRP National Cooperative Freight Research Program NCHRP National Cooperative Highway Research Program NHTSA National Highway Traffic Safety Administration NTSB National Transportation Safety Board PHMSA Pipeline and Hazardous Materials Safety Administration RITA Research and Innovative Technology Administration SAE Society of Automotive Engineers SAFETEA-LU Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (2005) TCRP Transit Cooperative Research Program TEA-21 Transportation Equity Act for the 21st Century (1998) TRB Transportation Research Board TSA Transportation Security Administration U.S.DOT United States Department of Transportation