Bridge Rating Practices and Policies for Overweight Vehicles (2006)

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.

16 As discussed earlier in chapter two, bridge load rating is a fed- eral requirement. The bridge load rating factor, as defined in Eq. 1, is an important index for the bridge’s condition in man- aging the entire United States bridge network. Bridge load rat- ing is currently practiced according to the AASHTO Manual for Condition Evaluation of Bridges (2000), which also refers to the AASHTO Standard Specifications for Highway Bridges (2002). The result of load rating for a bridge is its ca- pacity to safely carry vehicular load. Therefore, when a vehi- cle exceeds the legal weight limit of the jurisdiction and needs a permit to operate, load ratings of the bridges the permit ve- hicle planned to cross are often used for examination. Because load rating uses standard vehicle loads, such as the AASHTO HS and the H loads, the load rating results are directly useful only when the permit vehicle’s configuration is close to the standard load used. Otherwise, the typical approach of re- viewing the permit is to load the bridge with the permit vehi- cle and then determine whether the bridge can sustain the load. The latter is referred to as bridge evaluation for permit review, as discussed in chapter two. It was also noted that a number of different terms have been used in the practice and literature to refer to the bridge evaluation process. These terms are identified in chapter two. Although bridge load rating is guided by the AASHTO MCEB (2000), when applied to bridge evaluation for permit review there is ample room for interpretation and thus nonuniformity (this was discussed briefly in chapter two). With respect to this issue, this chapter presents relevant in- formation collected from U.S. and Canadian transportation agencies. Bridge load rating and bridge evaluation for permit review are closely related actions as already discussed. For many steps of the two processes and procedures the same concepts and quantities are used. Therefore, the survey for this synthesis study attempted to gather information on the state practices in both, and the results are presented and dis- cussed here. VARIATION IN EVALUATION AND RATING PROCESS It has been observed that there is a variation in the manage- ment of the bridge evaluation and load rating process among state-level highway agencies in the United States. This varia- tion may result in different bridge evaluation and load rating procedures, in terms of various factors such as the level of de- tail considered and software tools. This issue is observed in the survey and discussed next. Tables C4-1A and C4-1B in Appendix C offer an overview of the population of highway bridges with load rat- ings in the United States and Canada. Load rating is a re- quirement of FHWA for all highway bridges in the United States. The result of load rating a bridge is its safe live (ve- hicular) load carrying capacity with reference to standard vehicle configurations. They include the HS and H loads dis- cussed earlier. In addition, the AASHTO specifications (MCEB 2000; Guide Manual . . . 2003) also include other standard loads, Types 3, 3S2, and 3-3, as shown in Figure 8. The load rating result, as defined in Eq. 1, is the load carrying capacity of the bridge component as the difference between the rated member’s capacity and the total dead load effect, with all the safety factors included. There are two levels of load rating that are prescribed in the AASHTO MCEB (2000); the inventory and the operating ratings. The inventory rating refers to the “normal” load carrying capacity (or normally allowed load) and the operating rating to the maximum load carrying capacity (or maximum allowed load). The existence of load ratings for a bridge indicates that some information is available about the bridge’s capacity, although sometimes the load rating is estimated based on engineering judgment. When a permit vehicle is reviewed for a particular bridge, the existence of the load rating itself can mean that a detailed and quantitative analysis is possible for a relatively small amount of additional work, because some information is already available about the bridge. Table C4-1A shows that all the responding agencies have more than 60% of the bridges within their jurisdictions with a load rating, except for Massachusetts, North Dakota, Ohio, Puerto Rico, and Tennessee. As to what percentage of the bridges have an electronic model available, the dif- ference between the state-level agencies is much more significant, varying from 0% to 95%, as shown in Table C4-1A. An electronic model here refers to a model that can be repeatedly used, but requires minimal updating for some of the input data, such as corrosion-induced section loss, re- duced strength owing to aging, the loading vehicle, etc. When such a model is available for a bridge, the bridge evaluation for permit review can be readily done and the re- sults will be more consistent, compared with manual calcu- lations. These electronic models may use the concept of CHAPTER FOUR BRIDGE EVALUATION FOR OVERSIZE/OVERWEIGHT PERMITTING

17 girder line, 2-dimensional grillage analysis, 3-dimensional finite-element analysis, etc. When asked whether availabil- ity of such electronic models for the bridges has had an im- pact on the uniformity in permit review, 25 respondents said yes and 14 no. Of those agencies that answered yes, 23 explained why. It shows that more state-level agencies agree that electronic modeling is an effective approach in improving uniformity in permitting. Table C4-1B shows the same data as C4-1A, but for Canada. It can be seen that considerably lower percentages of bridges in the Canadian jurisdictions have a load rating. Also, many fewer bridges have electronic models available for repeating and updating load rating. Tables C4-2A and C4-2B show the responses of the agen- cies to the question of who provides the service of rating (i.e., bridge evaluation) when needed, for United States and Canada, respectively. The responses show mostly that state personnel undertake this function in permit review. This indi- cates that efforts to improve uniformity in this area can be ef- fective when mostly only state personnel are involved, because more resources are available to state personnel compared with other levels. FIGURE 8 Three other standard vehicle loads of AASHTO (Types 3, 3S-2, and 3-3) (MCEB 2000).

VARIATION IN EVALUATION AND RATING PROCEDURE The AASHTO MCEB (2000) has been used extensively in guiding bridge evaluation for permit review. This has also left ample room for the engineer to decide on many issues and aspects in the evaluation. In addition, the newly adopted AASHTO LRFR Bridge Design Specifications (2004) offers another alternative for bridge load rating and bridge evalua- tion for permit review. Table C4-3 displays the responses of U.S. transportation agencies to the survey regarding their practice on some of these aspects of permit review. It shows that only one of the states (Pennsylvania) is using the LRFR specifications, most states use LFR, and many use both the Allowable Stress Rating and LFR. As to which load rating level is used, it can be seen almost uniformly that the operat- ing level is used for permit review. The reasons for using this level are mainly as follows: (1) the operating rating allows higher loads, so that the probability of having a permit ap- plication approved can be maximized and (2) the operating level is rational for infrequent loads with higher certainty. Note also that Canada uses different specifications for load rating. Therefore, the differences between the United States and Canadian practices are more noticeable. Load Placement How to place the load on the bridge in bridge evaluation for permit review and how to determine the associated load dis- tribution factor is one of the most important elements affect- ing the result. Tables C4-4A and C4-4B contain the responses of the U.S. and Canadian agencies, respectively, regarding this issue. Seventeen U.S. agencies load only one lane with the permit vehicle, whereas another 15 load other lanes in ad- dition to the lane loaded with the permit vehicle. One agency uses both methods. For comparison, the Canadian agencies mostly use multiple-lane loading (Table C4-4B). Note that quantifying the probability of multiple vehicle presence on a bridge is still a subject for further research, because it has not been scientifically proven which way(s) is more appropriate. The right-hand portions of Tables C4-4A and C4-4B show the responses of the agencies regarding possible re- strictions for the permit vehicle on the bridge. These restric- tions are usually imposed on the vehicle to reduce stress and therefore the risk of failure, which in turn increase the prob- ability for the permit application to be approved. These restrictions include the loading position of the permit vehicle on the bridge, vehicle speed, whether other vehicles are allowed simultaneously on the bridge, and whether acceler- ation or deceleration is allowed on the bridge. The first three options have been used by most of the agencies that re- sponded (88%), whereas the fourth option is less frequently used (45%). In addition, other measures have been men- tioned to permit heavy loads: restricting traffic under the bridge to be crossed by the permit vehicle, restricting time of 18 day of travel, and altering the vehicle’s configuration to dis- tribute the load to more members. These restrictions all can change the bridge evaluation result for permit review. Computer-Aided Modeling Bridge evaluation for permit review and routine load rating now are largely done using computer software programs. Therefore, which program is used may have a strong impact on the result, although conceptually the differences should not be significant between the different programs. Table C4-5 shows the models and corresponding software pro- grams used by the U.S. agencies. In addition to the finite- element analysis (FEA), grillage, and girder line methods given in the questionnaire, the following methods were also mentioned in the responses: load testing and in-house meth- ods and programs. These results show that the girder line method is the most often used by the responding agencies. Therefore, if uniformity in analysis methods and software programs is a goal, more emphasis should be placed on this method and its associated software. Permit Screening Approach As discussed earlier, many U.S. state agencies take a two-step approach in permit review: (1) screening the permits into two groups, one requiring bridge evaluation and the other not, and (2) performing bridge evaluation if required. Various screen- ing concepts and approaches were cited in the responses. Table C4-6 summarizes the findings in this area. A large ma- jority of the responding agencies use comparison with the de- sign vehicle and/or acceptable axle spacing and axle weight for this screening. Other approaches are also being used including (1) comparison with the standard rating vehicles, (2) agency-specific formula (e.g., as used by Indiana), (3) comparison with the Federal Bridge Formula, and (4) com- parison with previously approved permits. The comparisons may be done using charts, maps, and/or computer programs. It is interesting to note that Kansas, Nebraska, and North Dakota use a rather unique approach that examines every bridge on the selected route, so that no screening is needed. It appears that the systems include electronic models for all the bridges in the jurisdiction. This approach is believed to be able to maintain a high level of uniformity, at least within the respective states. Bridge Condition and Material Properties Bridge load rating requires quantified estimation for bridge components’ material properties. When existing old bridges are involved, this estimation may not be uniformly done. Table C4-7 shows that approximately one-third of the U.S. agencies do not have specifications or guidelines as to how

19 bridge condition is taken into account in load rating or bridge evaluation for permit review, and seven reported that no specifications or guidelines are used for material property es- timation for bridge evaluation and load rating. This situation may have contributed to nonuniformity in bridge evaluation for permit review, because individual and possibly inconsis- tent decisions may have been made regarding these issues. Vehicle Gage Width The gage width here refers to the center to center distance be- tween dual wheel tires in the direction of the axle (see Fig- ures 1 and 2). Gage width is important in bridge evaluation because it directly affects the lateral distribution of the axle load over the bridge superstructure. In the case of beam-type bridges, the gage width directly affects the fraction of a wheel load to be transferred through the deck to the beams supporting the deck. The AASHTO vehicles used in the design of bridges are based on trucks that have a gage of 6 ft. Consequently, a gage of 6 ft is considered to be the standard gage width by virtually all permitting agencies in the United States. Gage widths that vary significantly from the standard 6 ft gage are often referred to as nonstandard gage widths. Although many OS/OW vehicles have standard or near standard gage widths, many do not, and typically the very heavy OS/OW vehicles (i.e., superloads) have nonstandard gage axles. In bridge evaluation for permit review, there is consider- able disparity between the states regarding the lateral distri- bution of the nonstandard gage axle loads. This is because there is no nationally recognized specification for the distri- bution of nonstandard gage axles. As seen in Table C4-8, the most frequently used specifications for lateral distribution is by far the AASHTO Standard Specifications for Highway Bridges (2002), followed by the AASHTO LRFD Bridge De- sign Specifications (2004) and the AASHTO Guide Specifi- cations for Distribution of Loads for Highway Bridges (1994). However, these specifications are all based on the standard gage axle and do not have specific provisions for the distribution of nonstandard gage axle loads. Tables C4-8A and C4-8B show the responses of United States and Canadian agencies, respectively, as to how they deal with this issue in permit review. Sixteen of the respond- ing agencies indicated that gage width is not taken into account in permit review at all, and those who do take into account gage width use the following methods or concepts: • Empirical distribution factors as functions of gage width, such as the effective width concept of McLelland (2003); • Lever rule; • A method cited in “A Rational Procedure for Over- weight Permits” (Bakht and Jaeger 1984); and • A method described in Bridge Analysis Simplified (Bakht and Jaeger 1985). It would clearly be valuable to compare these methods and to eventually develop a more widely acceptable method (that can be one or a combination of these methods) to have con- sistent practice. For multilane loads, columns 3 and 4 of Tables C4-8A and C4-8B provide additional details on this subject. Cranes actu- ally can be viewed as a special case of OW vehicles; however, many transportation agencies single out them for special treat- ment, as can be seen from these tables. Other Details in Load Rating Tables C4-9 and C4-10 exhibit the responses of the agencies to the questions regarding various details in bridge load rating. They include dead load distribution, span length definition, treatment of rebar cutoffs in concrete members, determination of dynamic impact factor, load effects considered as limit states, lateral load distribution factor, and inclusion of nontra- ditional additional loads and environmental factors. According to Tables C4-9 and C4-10, the following factors appear to be relatively uniformly treated: span length defini- tion, dynamic impact factor, and dead load distribution. The most nonuniformly treated factors are, in order of decreasing severity, environmental factors, bar cutoff, additional dead loads, limit state, and lateral distribution factor (with the gage length issue excluded). Consideration of the listed environmental factors varies from “not considered at all” to “as detailed as in design.” The same is observed for bar cutoffs. Bar cutoff here refers to steel reinforcement discontinued in concrete members. It causes a sudden reduction in load carrying capacity at the cutoff sec- tion. These types of sections may require checking in bridge evaluation as critical sections. As seen in Table C4-10, the most commonly concerned limit states are moment, and then shear. The most commonly used specifications for lateral load distribution factors by far is the AASHTO Standard Specifi- cations for Highway Bridges (2002), followed by the AASHTO LRFD Bridge Design Specifications (2004) and the AASHTO Guide Specifications for Distribution of Loads for Highway Bridges (1994). Local Bridge Evaluation Of the total numbers of bridge structures in this country, there are more locally owned bridges than there are bridges owned by state-level agencies. Also, local bridges have a higher percentage of their inventory with a lower load

carrying capacity. Therefore, the safety of most of the en- tire bridge population deserves adequate attention. How- ever, it is known that not all local bridges are maintained to the same condition level as state-owned bridges, for a vari- ety of reasons, including the demand on local bridges being relatively lower in terms of level of traffic, so that they do not have to be maintained to a higher standard and the re- sources available to local bridges are not as adequate. Table C4-11 shows the responses of the state-level agencies re- garding load rating for local bridges. Fifteen of the agencies do not know whether their local bridges are evaluated us- ing the same procedure used for the state-owned bridges, and four agencies of the 44 that responded believe that they are not. 20 SUMMARY As discussed in this chapter, load rating involves many details on which there could be a large variety of available approaches. These approaches can lead to different results. Conversely, the quantification of these differences has not yet been done to understand the nonuniformity thereby caused. In addition, the number of cases of permit review re- quiring bridge evaluation may be relatively small, compared with the number of permits issued without bridge evaluation, although the weights of those loads requiring bridge evalua- tion are much higher. Calibrating these various approaches along with the computer software programs used can be an effective approach to improved uniformity.