The Traffic Signal Network

Introduction

The Arizona legislature is currently entertaining two bills, either one of which would have unpleasant consequences for Arizonans if enacted. HB2277 (http://www.azleg.state.az.us/legtext/45leg/1r/bills/hb2277p.htm) and HB2278 (http://www.azleg.state.az.us/legtext/45leg/1r/bills/hb2278p.htm) are both designed as a response to issues related to photo enforcement.

HB2278 appears to get much of its verbiage from the model law put out by the National Committee on Uniform Traffic Laws and Ordinances (NCUTLO).

Besides the obvious issues with regard to due process, civil-rights, and other matters of that nature, both of these bills in general, and HR2277 in particular, will have a significant impact on traffic signal operations in Arizona if they become law.

Problems

What is the major concern? One of the changes to ARS 28-645 (3) "Red indication:" contained in HB2277 reads as follows (the modified text is in red caps):

(a) Except as provided in subdivisions (b) and (c) of this paragraph, vehicular traffic facing a steady red signal alone shall stop BEHIND THE LEADING EDGE OF THE FIRST CROSSWALK OR SAFETY ZONE LINE before entering the intersection and shall remain standing until an indication to proceed is shown...

This simple change has the following major consequences:

• Turn on red SAFETY will be negatively impacted
• Intersection clearance distances will instantly be increased

The NCUTLO (http://www.ncutlo.org) produces the Uniform Vehicle Code which is a tool used to help write state laws. The purpose is to keep uniformity between the states to avoid the chaos of each state treating things significantly differently. For example: California, Illinois, Utah, and Wyoming do not deviate in their definition of intersection sticking to the long-established standard of prolongations of the lateral curb lines or, if none, then the lateral boundary lines.

Although the definition of the intersection itself is not actually changed by this modification, the effect is the same. Do Arizonan's really gain anything by this change On the contrary, we have much to lose. For example, when approaching an intersection and attempting to turn on red it is quite often necessary to pull forward in order to have adequate vision. This change would make that illegal. Please read subdivision (b) of ARS 28-645 (3) to really grasp this:

(b) The driver of a vehicle that is stopped in obedience to a red signal and as close as practicable at the entrance to the crosswalk on the near side of the intersection, or if there is no crosswalk, then at the entrance to the intersection, may make a right turn but shall yield the right-of-way to pedestrians and other traffic proceeding as directed by the signal. A right turn may be prohibited against a red signal at any intersection if a sign prohibiting the turn is erected at the intersection.

It combined with the above wording would require a turn-on-red driver to turn from BEFORE the crosswalk. This, in turn, results in a DECREASE in intersection safety. An increase in accidents is often followed by a prohibition of the act that appears to be causing them. Prohibiting turn-on-red has the net result of a decrease in intersection capacity and efficiency - leading to more traffic problems and more pollution.

Traffic signal timing is affected by intersection characteristics. Yellow change intervals and all-red clearance intervals vary from intersection to intersection according to the needs at hand. One of the parameters is the intersection width. This change would lead to the inclusion of crosswalks in the calculations for signal timing. For example, a 10-foot crosswalk, 4-foot setback stopbar along would result in the effective intersection width being 28 feet more. In many cases, this distance is more like 35 feet on each side or an increase of 70 feet. This distance will need to be dutifully calculated and included in the yellow change and all-red intervals. Increasing these intervals affects traffic - and not always for the better. Longer yellow times can lead to driver's "pushing the yellow" as they become accustomed to the longer interval (e.g. learned disrespect for traffic control devices) Also, in essence, vehicles are now exposed to being in transit through the intersection for a longer period. It must be realized that this time must come from somewhere... that somewhere is the green time. Removing just a second or two from green time can have drastic effects on traffic flow resulting in loss of efficiency, backups, pollution, frustrated drivers (that often take chances), and (certainly) no improvement to the situation.

There are also a few other technical issues with this change:

• Traffic signal timing would need to be verified and adjusted every time the intersection striping changes.
• Photo enforcement detectors would need to be moved if the crosswalk and/or stop bar striping was modified.
• The photo enforcement detectors would be in direct conflict with in-roadway detection currently in place*

* it is typical for traffic signal loops to be placed so that they begin somewhere forward of the stop bar. Even under the new law, reality is that drivers would still pull forward in many instances. With detection behind the stop bar it would be quite easy for a vehicle to pull off of the detection - and if they don't get to turn into a gap on the cross-street the signal may never change (hence the reason detection is currently put where it is). Under the proposed law many existing loops would need to be ripped out and replaced by the two loops typically used by photo enforcement (with a new loop for the traffic signal placed behind them).

Solutions

So, what are the issues and how can they be fixed?

• Safety Status
• Proper Criteria & Operating Principles
• Driver Education
• Enforcement Costs

Firstly, much noise has been made that Arizona is ranked as one of, if not THE, worst location for red-light fatalities. Well, let's look at that more closely. The study often cited was performed by the Insurance Institute for Highway Safety (a private non-government institute). This study lumped together seven years of data (to get a large enough number). The details of the study are not reported - just the ranking. And even looking at the IIHS press release can be misleading unless the viewer notes that the data is seven years lumped together. Breaking the data back down to more normal numbers means that Mesa's numbers, for example, are 1.1 deaths per 100,000 population per year. The ranking will stay the same regardless, however, the reality looks different with this perspective. The true rate is significantly better than in "the good old days" of years gone by and than that of developing countries. Improvement can, and should, certainly still be made. Another point to consider is that many of the locales ranked in the study have significantly lower speed limits and narrow roads - it is more difficult to be killed at lower speeds... the real irony being that the fine standards to which Arizona roads are typically built leads to faster speeds and, consequently, better opportunities to die in an accident.

Even one death IS too many. However, we must not give up our rights and cause ourselves other problems based on false assumptions about the seriousness of the situation: we must know reality to deal with it. Sadly, the reality is that many locales do not look at their own statistics to really know which locations to target. Quite often the installation and operation of photo enforcement is handled solely by the police agency with no before-hand statistical data used in the installation decision-making process.

Contained within the NCUTLO model law (http://www.ncutlo.org/autoenforce622.htm) are some good concepts that should be followed:

• Only use automated enforcement at locations that have been shown to have a violation-related accident problem.
• Do an engineering study and make sure the location meets safety standards before implementation (automated enforcement is not a means of "combating deficiencies in roadway design or environment"). 
• Start a public information campaign BEFORE implementing photo enforcement
• Do NOT base vendor fees on a percentage of revenue
• Continue the public information campaign all during time when system is in place
• Evaluate the program and discontinue it if it hasn't helped

Many of these principles are not being followed at present. There are agencies with contracts that are based on a quota and there are even contracts that dictate that signal timing cannot be changed. Quite often locations are chosen not based on specific accident history - rather they are chosen as "major intersections" or for political or other reasons. These and other factors really make it look like a revenue issue and not a safety issue. Safety must be the focus - it is not really about enforcement and certainly not about revenue. Sound engineering principles must drive the implementation of automated enforcement to assure that it is 1) necessary, and 2) functional. If a program breaks-even great. If it doesn't but it creates a safer environment that fact can easily be touted and will most-likely be supported by the public. However, a program that has serious issues that make it appear to be revenue related should be avoided and will be hard-pressed to find public support.

Education should be a key component to any of these programs. Not enough is being done in this area. Drivers should not have to go to court to find out the legal definition of the intersection (and nor should the definition be changed for previously cited reasons). To that end, advertising is a good idea as well as is another key obvious component: paint the intersection line. Some jurisdictions are doing this already. It makes it plain to the motoring public what the reality is.

Photo enforcement is expensive and it gets more costly when it becomes necessary to use two cameras. The simple solution to that problem is to require front license plates. Arizona is one of only a handful of states that do NOT require front license plates (those that only require front plates currently are: Alabama, Arizona, Arkansas, Delaware, Florida, Georgia, Indiana, Kansas, Kentucky, Lousiana, Michigan, Mississippi, New Mexico, North Carolina, Oklahoma, Pennsylvania, South Carolina, Tennessee, and West Virginia according to http://mrtraffic.com/frontplate.htm). With a front plate the cost of photo enforcement immediately declines. This would be a simple change that would take two years to implement (with vehicle renewals). There are costs associated with it, however, it is a) not unreasonable, and b) does not trample individual rights.

Various agencies are finding that the word is clearly out that photo tickets can be ignored unless the police follow up on them. The cry has gone out that this is costing the programs significant dollars. This does not need to be the case. The simple solution involves three steps all being paid for by the offender:

1. Send the ticket by regular mail - if the driver responds, OK - otherwise proceed to the next step
2. Send the ticket by registered mail - if the driver responds, OK - otherwise proceed to the next step
3. Send out a process server

This is not difficult and is there any reason why the offender can't and shouldn't be expected to pay for the increased cost based on their decision to "ignore" the ticket? The burden then truly is on the "bad guys" but NOT at the expense of civil liberties of due process. Another point to make is that the tickets themselves could be improved as well. For example, I've seen tickets that show a few small not-very-clear photos on them that don't include the all-important shot of the vehicle in the intersection (from the front). In one instance in one rear photo it appeared as though the vehicle had stopped in the crosswalk (and that's what the driver of the vehicle was claiming happened). The front picture that was taken had been blown up to just show the driver's face. In order to prevent issues with "who else was in the vehicle" the windshield in the all-important front vehicle shot could be blacked out electronically. This should help convince more people that they were legitimately ticketed for their offense.

A Real Problem

This problem has caught many drivers. Although legally they do not have a leg to stand on (i.e. they did cross into the intersection AFTER the signal was red), the reality is that their expectations were violated. This issue has resulted in many technically correct but seriously flawed tickets being issued to the tune of thousands of dollars. Many Arizonans have been trapped and abused by the implementation of automated enforcement.

At 45 MPH a vehicle travels at 66 feet per second. As currently defined, the intersection is the prolongation of the lateral curb lines (extend the curb lines along each side of the street through the intersection to draw a box). The crosswalk and stop bar are typically behind this point. The distance from the intersection line to the stop bar is in the neighborhood of 20 feet on a typical through movement. At 66 feet per second this takes around 0.3 seconds to traverse. A red-light camera typically provides a 0.1 second minimum delay from when the signal turns red until it begins snapping photos. The City of Mesa, for example, adds an additional 0.3 seconds to this time. Therefore, in most cases for drivers that believe the stop bar is the intersection, they will not be ticketed because if they cross the stop bar just before the end of the yellow and are doing around the speed limit, they will pass into the intersection before the camera starts taking pictures.

Left turn arrow locations are different. Quite often the stop bar will be significantly back from the intersection line for the left turn lane. Intersection geometry and dual-left turns make this a necessary reality so that vehicles don't collide (i.e. if vehicles were waiting at the intersection line cross street left turning vehicles would not be able to clear them. This distance can easily be in the 35 foot range at many locations. At a typical left turn approach speed of 15 to 30 MPH (22 to 44 feet per second) this distance takes 0.5 to 1.6 seconds to traverse. This means that the left turning driver that thought the stop bar was the intersection line will be ticketed.

Even if a driver knows that the stop bar is not the intersection still has a problem. Due to the geometry of the intersection it is quite reasonable to use the stop bar as a guide for stopping. To stop before the intersection may be legal, but it would still expose the driver to cross street left turners. The options are scant: 1) back up (which is illegal) 2) stay in place and hope the cross street drivers clear, or 3) proceed through (and be ticketed if automated enforcement is in place. Of course, this decision is mostly made as the driver approaches the yellow indication (think fast).

The City of Mesa (and possibly other locales as well) currently violate driver expectations when cameras do not take into account the realities of intersection geometry. This could easily be fixed without legislation. Rather than set the camera to a fixed 0.4 second delay, vary the delay according to the distance from the stop bar to the intersection as necessary on a location-specific basis. This will assure that only those that really were intentionally running the light will be ticketed and will clear up the current innapropriate situation.

Take Action

Please contact the legislature regarding this issue and contact your local City Council as well (Be aware that Cities comment on pending legislation and have lobbyists).

Below are the email addresses and phone numbers of the sponsors of these bills:

HB2277

HB2278

If you are a resident of the City of Mesa, below are the email addresses of the Mayor and City Council:

[Home] [E-Mail]

Copyright © 2000-2001 The Traffic Signal Network. All Rights Reserved