The Impact of Load Rating Methods on Federal Bridge Program Funding
FHWA May 2005 (revised February 2006)
Beginning with the April 2005 data collection for the National Bridge Inventory (NBI), the FHWA allowed the inventory rating and operating rating to be reported as a Rating Factor (RF) using either the Load Factor Rating (LFR) method or Load and Resistance Factor Rating (LRFR) method. FHWA's memorandum dated March 22, 2004, Attachment 1, outlined these changes. Prior to this change, the LFR method using MS loading (HS metric equivalent) was the national standard for computing inventory and operating ratings reported to the NBI.
The change provided the ability to report Allowable Stress Ratings (ASR), LFR or LRFR for both operating and inventory ratings (NBI Items 64 and 66) by a RF rather than a tonnage. Also new was the concept of using different loadings depending on the method of rating, i.e. MS for Allowable Stress (AS) and LFR vs. HL-93 for LRFR. The possible affects these changes have on the federal bridge program, specifically funding apportionments, will be reviewed.
It's important to note that only eligible structures determined to be deficient, i.e. structurally deficient (SD) or functionally obsolete (FO), are considered in the process for apportioning Highway Bridge Program (HBP) funding. A sufficiency rating (SR) for each structure is determined for program eligibility purposes. Deficient structures, as defined in attachment 2, with an SR value less than 50 are eligible for replacement, 50 to 80 are eligible for rehabilitation, and above 80 are not considered eligible. Other eligibility rules also apply, i.e. the structure must be of bridge length and had not been constructed or had major reconstruction within the past ten years. Bridges must be deficient and eligible in order for their deck area to be considered as the basis for the HBP funding apportionments.
The inventory rating, NBI item 66, is the only load-rating item that has the potential to affect funding levels. Item 66 is a factor in determining the calculation of NBI item 67, structural evaluation. It is also a factor in calculating the reduction for load capacity, B, in the S1 portion of the SR formula. Many States use the operating rating, NBI item 64, to determine if the structure can carry legal loads and make posting decisions, but again, the operating rating does not affect the HBP funding levels.
A recent NCHRP project 20-07/task 122 study by Dr. Dennis Mertz of a small sample of bridges reported that LRFR average about 7% higher than LFR for design-load inventory ratings. Our findings indicate that this would cause less than a 2% reduction in national eligible deck area on deficient bridges if all NBI inventory ratings were suddenly changed to LRFR. Other variations in LRFR vs. LFR, and the resultant impact on eligible deck area on deficient bridges, were evaluated. The findings are presented in the attachments. The FHWA does not intend to require the States to re-rate the large inventory of older structures that have already been rated by ASR or LFR with LRFR methods, but rather to allow the States to re-rate existing structures with LRFR per their own policies. Only structures designed or reconstructed using LRFD would be required to be rated using LRFR.
The FHWA March 22, 2004 memorandum revised the NBI Coding Guide to allow for the reporting of rating factors for ASR and LFR using a MS-18 loading and LRFR using a HL-93 loading. The notional highway loading HL-93 for LRFD and LRFR was developed to provide a more uniform safety factor for structures over various lengths, be more inclusive of AASHTO and State legal loads, and to include legacy exclusion trucks, but this loading is not expressed in tons. The result of a LRFR using HL-93 loading is a RF and conversion of this RF to the current HS loading ton value is not easily done.
The concern of having structure inventory load ratings reported in a uniform manner using multiple design and rating methods each using a different loading function is addressed with the use of rating factors based on the design method and a standard loading for the method of design and rating. Logically, structures designed with allowable stress or load factor, should be rated with LFR using the MS-18 loading. Structures designed with LRFD should be rated using LRFR and the HL-93 loading. This relation between design method and rating method, and their corresponding loads, is not currently an AASHTO or FHWA requirement.
We have historically assigned a tonnage value to a MS-18 or HS-20 load rating performed by the LFR method. The RF is easily determined for these loadings by dividing the rating tonnage by 32.4 tons for MS-18 loading or 36 tons for HS-20 loadings. The result of a LRFR using HL-93 loading is already a RF. The resultant RF for the structure will therefore be determined by a similar methodology and loading function as it was designed for. This will allow for a uniform reporting procedure based on how well the structure is performing in comparison to the current standard for the given design methodology.
The design-load rating values in NBI items 64 and 66 are directly affected by the rating methods; however, only item 66, inventory rating, could affect a bridge's deficiency status and the SR. There are two locations in the NBI coding guide where NBI item 66 is used. The first is item 67, structural evaluation. NBI item 66 along with Average Daily Traffic (ADT) is used in a table to determine a possible value for item 67. The value of item 67 is also based on the coding values of items 59, superstructure, and 60, substructure, or 62 if the structure is a culvert. The lowest of the coding values of these items, or from the table using item 66 and ADT, determines the value of item 67. Item 67 is used directly in determining a small part of the "S2" component used in the calculation of the SR. The importance of item 67 is that if its value is a 3 or lower, the structure would be considered deficient (see attachment 2). If item 67 is a 3 or less due to values in the table, the highest possible SR is 78, making the structure eligible for rehabilitation using HBP funds.
The inventory rating/ADT table used in coding item 67 could easily be modified to be based on RF for clarity rather than the current use of tonnages. But, as used in the FHWA 2004 memo, converting the RF to a pseudo number by multiplying by 32.4 to use in the table will yield the same rating for item 67 as if we had converted the table to RFs. (See Attachment 5)
The second use of the inventory rating (item 66) is in calculating the reduction factor "B" of the "S1" component in the SR formula. See appendix B of the coding guide.
The base equation for this calculation is:
B = (1-RF)1.5 x 60.0 where RF = rating factor
In the past, we have expressed this equation in two different forms based on the whether the ratings were to be reported in HS-20 or MS-18 loadings.
Replacing inventory rating (IR) with either:
IR = RF x 36 for HS loadings or
IR = RF x 32.4 for MS loadings
yields the following equations that have been used in appendix B of the coding guide:
B = (36.0 - IR)1.5 x 0.2778 where IR is NBI item 66 Inventory Rating in US tons
B = (32.4 - IR)1.5 x 0.3254 where IR is NBI item 66 Inventory Rating in metric tons.
In the future, for clarity, perhaps this equation should express item 66 in RF rather than the current IR. Presently, as indicated above, the RF for item 66 can be multiplied by 32.4 to yield a pseudo value for determining "B". This pseudo number does not represent a real tonnage load but just a number needed to allow the current programs to operate correctly. As stated above, the end result, i.e. the coding of item 67, the "B" value, and the SR value, are the same whether you use the actual RF or the pseudo number.
The important concept to remember is that the "B" value, and hence the SR, are being determined based on how well the structure is performing in its current inspected condition in comparison to the rating standard, i.e., the appropriate design load.
By rating a bridge using LRFR instead of LFR, the RF for a given structure may change. The concern of how this change might affect a structure's deficiency status and SR, thereby influencing individual States HBP funding apportionments, has been raised.
As previously stated, only deficient and eligible bridges are considered in the apportionment process. Once a structure is considered deficient, its SR is used to determine if it is eligible for funding. Structures with a SR above 80, under construction or reconstruction, or that have been built or reconstructed in the last 10 years are not eligible. Structures with a SR below 50 are eligible for replacement and structures with a SR between 50 and 80 are eligible for rehabilitation. The deck area for this group of eligible and deficient structures is divided into four groups: on-system and off-system with SR between 50 and 80 and on-system and off-system with SR below 50. The three-year average bridge construction unit cost data as reported by each State is then used to determine a total cost to replace/repair the structures. Different costs are used for the on-system versus the off-system, and for replacement versus rehabilitation. Unit costs for rehabilitation are taken as 68% of the replacement costs. This "needs" calculation (deck area (times) unit cost) is generated for each state. A ratio of the state needs vs. the national needs (sum of all state needs) is determined and that becomes the state's apportionment factor for calculating their share of the total HBP funds available in a given year. Some corrections are used such that no State's share exceeds 10% or is under 0.25% of the total national fund.
The question of what would happen to the eligible deck area on deficient bridges if States were to re-rate all their bridges from LFR to LRFR will now be reviewed. First the FHWA suggests that only those structures designed or rehabilitated using LRFD should be rated using LRFR with HL-93 loading. This should alleviate most concerns over sudden changes in deck area on deficient bridges, as any changes would occur over an extended period of time, as LRFD becomes the standard design methodology for all States.
To understand what could happen if states were to re-rate all their structures using LRFR, we varied NBI item 66 for the complete bridge inventory to determine how susceptible the eligible bridge area is to change (See attachment 3). The Bridge Management Information Systems Laboratory at the FHWA Turner-Fairbanks Research Center performed 9 NBI data runs where they varied item 66 (inventory rating) to determine the affects on eligible area on deficient bridges. The data runs were made by varying item 66 from plus 20 percent to minus 20 percent, in 5 percent increments. Based on this analysis, several charts were produced. Three bar charts show the relative percent change in eligible deck area for all highways, federal-aid system and off system bridges for the three groupings of SR values: SR = 80; SR < 50; 80 = SR = 50. A fourth chart (shown first) was developed for the analysis with SR = 80 showing a factored deck area where the full deck area of replacement structures is used, but only 68% of the area of rehabilitation is used for structures eligible for rehabilitation. This will capture the affect of the difference in values (unit costs) assigned to replacement vs. rehabilitation.
As a check, the Office of Bridge Technology examined the sensitivity of deck area on deficient bridges to variations in NBI item 67. This is the primary item affected by NBI item 66 that could change a structure's deficiency status. It should be noted that if a structure becomes deficient due to item 67 becoming a 3 or less based on the table, the maximum SR the structure could have would be a 78, making the structure eligible for funding. We divided these structures into two groups: See attachment 4.
Structures that if re-rated higher would no longer be deficient. (Item 67 changes from a 3 or lower to a 4 or higher)
The category of deficient bridges where NBI item 67 is a 3 or less, and item 67 is the only item causing the structure to be deficient so that raising it to a 4 would cause the structure to no longer be deficient, includes only 3.6% of the national eligible deck area on deficient bridges. Individual state percentages vary, indicating that the impact on funding would also vary. It must be emphasized that these percentages are based on a somewhat unrealistic scenario, i.e. all States chose to re-rate their entire inventory with LRFR, and all the bridges in this category rated high enough so that they were no longer deficient. Based on research by others comparing LFR to LRFR, it is known that such an assumption is not completely realistic.
Attachment 4 provides statistics based on the NBI. Columns 4 and 5 provide the total number and area of eligible deficient bridges by state. Columns 8 through 10 provide data on bridges that are deficient due only to item 67 being under a 4. If any of these bridges were to be re-rated higher using LRFR so that item 67 became a 4 or higher the corresponding bridge deck area would no longer be considered in the apportionment process.
Structures that if re-rated lower could become deficient and eligible. (Item 67 changes from a 4 to a 3 or lower)
The last 3 columns of attachment 4 display information on the number of structures that are not deficient and item 67 equals a 4. It is assumed that if any of these bridges were to be re-rated lower using LRFR and item 67 drops to a 3 or lower, the bridge deck area would become eligible. In practice the 10-year rule would prevent a number of these structures from becoming deficient. If we assume a State was to re-rate their bridges and all this groups' area would become eligible, the total deck area on deficient bridges could rise by 9%. The likelihood of all this area becoming eligible is remote. Further studies of samples of bridges indicate that LRFR will give about a 7% higher rating than LFR, making this case even more unlikely.
A recent research study of a small sample of bridges reported that LRFR average about 7% higher than LFR for inventory ratings. Using the charts for +5% and +10% (attachment 3) increase in item 66 for the factored bridge area category, the average change on a national level for area on deficient bridges is -1.7%. The maximum and minimum values range between -0.1% to -5.8% for the individual States. Earlier studies indicated that LRFR might yield lower inventory rating results, as an example looking at the charts for a 10% lower rating for item 66 results in a national increase of eligible deck area on deficient bridges of 5%. Considering the small magnitude of these changes, it is unlikely that the actions of a few states would affect the funding of other states. Full implementation of LRFR is likely to occur gradually, as LRFD becomes more common, making the changes in deck area on deficient bridges even more difficult to detect.
For structures that were designed with allowable stress or load factor, the continued use of LFR reported in either a tons rating or a rating factor based on MS-18 would be considered the standard practice. If a State desires to rate a structure designed with allowable stress or load factor using LRFR, the corresponding HL-93 loading must be used. For structures designed or reconstructed using LRFD, then LRFR using HL-93 loading is to be used to rate the structure. For structures that use load tests to determine a load rating for item 66 those loads will be reported in rating factor based on a MS-18 truck, even though the actual load test was likely performed with other than an MS-18 truck configuration.
Operating ratings (NBI item 63) offer a different set of issues and problems. Once LRFR inventory and operating ratings, based on design loads (HL-93), are determined for possible screening of allowed loadings, legal and permit load LRFR ratings for specific trucks like the AASHTO type 3 trucks can be used for posting and permitting (see LRFR specification flow chart 6-1). The tonnage for these specific trucks can be determined and used for posting and permitting. It is the intent of the FHWA to collect only inventory and operating ratings based on the design loadings, which in the case of LRFR would be HL-93, and for AS or LFR would be MS-18.
For structures that were designed with allowable stress or load factor, the continued use of LFR reported in either a tons rating based on a MS loading or a rating factor based on MS-18 loading would be considered the standard practice. We will convert LFR using MS-18 loadings to a rating factor by dividing the tonnage rating by 32.4 to determine a rating factor. If a State desires to rate a structure designed with allowable stress or load factor using LRFR methods, to report in the NBI, it may do so provided that HL-93 loadings are used. Structures designed with LRFD using HL-93 are to be rated and reported to the FHWA based on LRFR methods using HL-93 in a rating factor.
- Attachment 1: March 22, 2004 FHWA memorandum
- Attachment 2: Method for determining if a structure is deficient
- Attachment 3: Results of data analysis by BMISL
- Attachment 4: Results of data analysis by Office of Bridge Technology
- Attachment 5: Proposed revisions to Coding Guide Item 67 table to allow rating factor