Utilities and Roadside Safety (2004)

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.

18 4 Strategies Don L. Ivey and Charles V. Zegeer The primary objective addressed by strategy is to maximize the benefit to society forevery action and expenditure taken by state departments of transportation (DOTs),local highway agencies (HAs), and utility companies (utilities) in a collision reduc- tion program. A secondary but important objective is to institute a program that will pro- vide state DOTs, local HAs, and utilities with the best defensive position relative to potential litigation. Strategies and analytical methods to maximize these benefits have been developed and used by some state DOTs and a few utilities and municipalities with good results (see Chapter 5). Examples of these are the approaches of Zegeer and Parker (1), that of Griffin et al. (2), and the method presented in the current AASHTO Roadside Design Guide (RDG) (3). Each of these approaches has advantages. The Zegeer approach relies on only three fac- tors: traffic count, lateral distance to poles, and pole density to predict accident frequency. It is statistically derived but admittedly has limited predictive power. The Griffin approach is potentially more powerful statistically but is somewhat more complicated to apply. The RDG is a benefit–cost approach, used most commonly to evaluate the potential dan- ger of roadside objects and how that compares with the application of guardrails or crash cushions. The authors have not been insulated from the field of litigation and have developed a clear understanding of the strategies of both the plaintiff and the defendant over the past three decades. As part of these analyses and efforts to achieve an optimum approach, every effort has been made to find and take advantage of the strengths of various approaches while respecting the justifiable objectives and economic constraints of utility companies. In so doing, the following objectives were considered: 1. Prevent the recurrence of a fatality or injury at sites where collisions have already occurred. 2. Prevent the occurrence of a fatality or injury at sites where collisions are likely to occur. 3. Save the utility maintenance funds. 4. Put a utility in the best position to defend the clearly random collision. (This is poten- tially a way to save the stockholders and customers of a utility millions of dollars.) Don L. Ivey, Texas Transportation Institute, College Station, TX 77843. Charles V. Zegeer, North Carolina Highway Safety Research Center, Bolin Creek Center, 730 Airport Road CB-3430, Chapel Hill, NC 27599-3430.

Perhaps the term “clearly random collision or event” is not familiar. It is this event and the ability to define it that will provide a utility, local HA, or state DOT the best legal defense against litigation. One way of defining the clearly random event is that it is everything outside the realm of the predictable. Predictive equations have been devel- oped that are strong enough to make cost-effective site selections. If utilities, local HAs, and state DOTs act in an appropriate way in prioritizing and treating the predictable, then a strong defense can be laid not only against the clearly random but also against the lower-priority levels of the predictable. The result is the following three-path approach: best offense, best bet, and best defense (4). BEST OFFENSE This is the most obvious of the approaches and historically the most frequently used, residing firmly in the realm of common sense. It is improving safety where an atypical number of collisions have already occurred. It will work toward Objective 1, prevent- ing the recurrence of a fatality or injury at sites where multiple collisions have already occurred. What is required is for a utility to know where collisions are occurring. There are two practical approaches. The first and most direct is to make arrangements with the appropriate law enforcement agency or agencies to secure copies of all collisions involving a part of the utility’s physical plant (e.g., utility poles). Those collisions are then located to determine the facility sustaining the damage. Usually at least 3 years of accident data are necessary to begin determining the most susceptible sites, but some- times a congregation of crosses placed by survivors or other observations in the field can be clues hard to misinterpret. The identified locations of a pole or poles atypically exposed can then be prioritized for movement or treatment. This is the most immedi- ately visible and most obviously effective portion of the safety program. Such a pro- gram was designed by Mak (5) and successfully applied by Jacksonville Electric Authority in 1989. However, this approach suffers from requiring unsupportably costly collisions for definition (i.e., it is reactive instead of proactive). It is a part of the pro- gram that should take priority early and gradually be reduced in importance as these obvious exposed areas are changed. Note that this approach will also help accomplish Objectives 2, 3, and 4. BEST BET In this phase of the program, pole lines and roadways are prioritized by statistical algo- rithms that can be applied before an accident history develops. Zegeer and Parker have developed prediction equations and data useful in prioritizing pole lines of significant length. Good et al. (6) have also developed useful relationships. These relatively simple equations rely on traffic volume, pole offset, and pole spacing to predict where the prob- ability of pole collisions is greatest. They are based on a comprehensive database from 1,534 roadway sections covering 25,193 roadway miles. The sections are in Michigan, North Carolina, Washington, and Colorado. Six to 10 years of accident data were required for each section. The analysis included more than 9,600 utility pole accidents. A major accomplishment was the development of a regression model (7) to predict utility pole accidents: where Accidents/mi/yr = number of predicted utility pole accidents per mile, ADT = annual average daily traffic volume, Accidents/mi/yr (ADT) + 3.54 10 density offset = × × − − − 9 84 10 0 045 2 0 6. ( ) ( ) .. Strategies 19

density = number of utility poles per mile within 30 ft (10 m) of the roadway, and offset = average lateral offset of the utility pole from the roadway edge on the section. Because pole collisions are generally low-probability events, the power of these algo- rithms to make accurate predictions is limited. Thus, this prioritization scheme should probably be only one of the controlling factors dictating change. It might be especially help- ful in concert with right-of-way expansions or roadway widening (i.e., DOT improvement projects). For example, if a DOT project allowed movement of a pole line from 10 ft behind the curb to 18 ft, there is a good probability the money for utility movement would be bet- ter spent elsewhere. Thus, a utility could propose, on the basis of statistical probability, that a higher-priority section of poles be moved or treated with the funds that would have been expended on the 10-ft to 18-ft project (e.g., where poles could be moved from 2 ft to 10 ft). Further, when a given pole line shows a high priority for change, that occasionally could be used by a DOT to justify the acquisition of more right-of-way. Note that this best bet approach will apply directly to Objective 2 and will help accomplish Objectives 3 and 4. Finally, something else should be accomplished while saving lives and limbs. Safety funds should not be dissipated on frivolous lawsuits. The final approach will be likely to prove a great frustration to plaintiff attorneys with unjustifiable lawsuits. BEST DEFENSE In the courthouse, a second legally damaging condition for a pole line, right behind a sig- nificant accident history, is failure to meet the recommendations of the RDG (3). This is already true for state DOTs, counties, and cities. It is likely to become true for utilities as the aforementioned governmental entities take the logical steps to share the responsibil- ity for roadside safety. This has been true even in cases in which the degree to which non- compliance with the RDG recommendations is slight. In Arizona, the city of Mesa was recently held accountable for a drainage structure 15 ft from the traveled way, while the RDG recommended 17.5 ft. A way of decreasing the liability for letter-of-the-law diver- gences from RDG recommendations is as follows: 1. Document the areas, pole lines, and individual poles that were originally placed or came to be placed in conflict with the clear zone recommendations of the RDG. 2. Use the physical characteristics of these sites to calculate the percent compliance (PC) value with the RDG. Interpret the PC value to secure a priority number (PN). Note the relationship between PN and PC with the RDG can be derived to achieve the most productive priority listing by using lateral encroachment predictions and relative risk relationships. 3. Schedule modification of sites according to the PNs. 4. Perform safety treatment of a reasonable number of the highest-priority sites each year. (Some utilities have found a cost-beneficial investment of $100,000 per year will yield effective progress.) In this way, if an area is in reasonable compliance with the RDG clear zone (e.g., there is a 15-ft clear zone instead of the recommended 17.5 ft), it will show up as a very low priority for treatment and thus place the state DOT, the local HA, and the utility in a good defensive position if one of these sites is subject to the rare and unpredictable random collision. Note that this third strategy pursued in concert with the first two will clearly accomplish Objective 4. The following is a simplified approach that was recently implemented by Lafayette Utilities System (see Chapter 5 for more detailed information). Step 1. Continue to monitor collisions with utility structures to determine whether sites are disposed to repeated collisions or are simply subject to the purely random collision that is unlikely to be repeated. 20 Utilities and Roadside Safety

Step 2. Apply predictive analyses to heavily traveled thoroughfares to determine the relative probability of collisions in selected areas or sites. Step 3. In the areas and sites that are prioritized by Step 2, determine the relative degree of consistency with the recommendations of the RDG. Step 4. Make safety modifications based on Appendix A to the top 10 sites each year. REFERENCES 1. Zegeer, C. V., and M. R. Parker, Jr. Effect of Traffic and Roadway Features on Utility Pole Acci- dents. In Transportation Research Record 970, TRB, National Research Council, Washington, D.C., 1984, pp. 65–76. 2. Griffin, L. I., III, R. J. Flowers, and G. E. Miller. Use of Empirical Bayes Procedure to Identify Loca- tions on the Texas Highway System That May Be Represented in Ran off Road Accidents. Texas DOT, Austin, Sept. 1993. 3. Roadside Design Guide. AASHTO, Washington, D.C., Jan. 1996. 4. Transportation Research Circular E-C030: Utility Safety: Mobilized for Action and State, City, and Util- ity Initiatives in Roadside Safety. Presentations from TRB Committee on Utilities (A2A07) from the 79th Annual Meeting of the Transportation Research Board, Washington, D.C., April 2001. gulliver.trb.org/publications/circulars/ec030/ec030.pdf. 5. Mak, K. K. Design Considerations for Safety Demonstration Sites. Texas Transportation Institute, Texas A&M University, College Station, Jan. 1989 (revised Sept. 2000). 6. Good, M. C., J. C. Fox, and P. N. Joubert. An In-Depth Study of Accidents Involving Collisions with Utility Poles. Accident Analysis and Prevention, Vol. 19, No. 5, 1987, pp. 397–413. 7. Utilities and Clear Zones: Analyses, Observations and Misconceptions. Presented to the Com- mittee on Utilities, TRB, Washington, D.C., Jan. 1998. Strategies 21