U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
|Toni Doolen, PhD|
School of Mechanical, Industrial, and Manufacturing Engineering
Oregon State University
|State||Members and Titles|
|Oregon||Benjamin Tang, P.E., Br Preservation Manager
Steve Soltesz, Research Coordinator
Dawn Mach, Bridge Fin. Analyst
Holly Winston, Sr. Local Bridge Standards Engineer
|FHWA||Mary F. Huie, Highways for LIFE, Program Coordinator
Tim Rogers, P.E., Division Bridge Engineer
Nat Coley, Asset Manager
|California||Paul Chung, Sr. Bridge Engineer|
|Iowa||Ahmad Abu-Hawash, Chief Structural Engineer|
|Minnesota||Kevin Western, Bridge Design Engineer|
|Montana||David Johnson, Bridge design Engineer|
|Texas||Courtney Holle, Transportation Engineer|
|Utah||Daniel Hsiao, P.E., S.E., Sr. Project Manager|
|Washington||Bijan Khaleghi, Design Engineer
DeWayne Wilson, Bridge Management Engineer
8 states + federal highways
Account for characteristics of bridge project, e.g. bridge length, complexity, road user characteristics, environmental requirements, traffic levels, existing levels of congestion, and construction site attributes
User-friendly and flexible to accommodate a range of construction situations, transparent as to the method of calculation, and customizable to maintain future relevance.
Engineers who can use the tool to create detailed estimates for recommendations.
TAC team members along with research team developed a comprehensive list of criteria that are relevant to the decision of when to use ABC tools/methods for a project. Each criteria was defined and sub-criteria were defined, as appropriate.
Key Resources on AHP
Saaty, T.L. (1980) The Analytic Hierarchy Process, McGraw Hill.
Saaty, T.L. and Vargas, L.G. (1984) .Comparison of eigenvalue, logarithmic least squares and least squares methods in estimating ratios., Mathematical Modelling, 5, 309-324.
|Criteria||Direct Costs||Indirect Costs||Site Constraints|
The Copano Bay Bridge: The Copano Bay Bridge replaces the existing causeway on SH 35 at the mouth of Copano Bay. The bridge connects the cities of Rockport/Fulton and Lamar, on the Gulf Intracoastal Waterway. Copano Bay is home to oyster colonies and migratory birds, attracting birdwatchers year-round. Two peninsulas frame the bay opening, limiting ROW and dictating phased construction. The bridge is 11,010 feet long, with a 129’ wide and 75’ tall navigation channel. The existing structure suffers severe corrosion from marine exposure, such that some piling members have failed and required extensive repair. As such, providing corrosion protection – in the form of high-performance concrete, stainless reinforcing steel, and cylinder pile foundations – was of high importance. The superstructure is 100’, 120’, and 150’ long prestressed concrete girders. A majority of the piers consist of cast-in-place caps on trestle piles, with the tallest piers around the navigation channel being CIP bent caps on CIP columns and waterline pile caps. Contractors may elect to propose precast bent caps as alternate construction, thus reducing the duration of construction activities over open water.
Existing Bridge is on Clear Creek, Gulick Lane
Existing Bridge length: 29ft steel girders on concrete vertical abutments
The bridge is on a rural local road.
Detour length: 1 mile
The new bridge will be 80-100 ft in length
The MMS (Moveable Scaffold System) underslung modular deck forming , span by span cast in place self launching form traveler, when there is insufficient room for casting yards or transport of segments is improbable. Photo Courtesy of Harsco Infrastructure Americas.
The first tab is associated with constructing a decision hierarchy. In this tab, the user has access to all necessary functions to support loading, saving, and modifying a decision hierarchy. The user has the option to disable a decision category either temporarily or permanently for every hierarchy. The second tab is associated with conducting pairwise comparisons. The user can save the state of an analysis at anytime and later return to that specific position, without losing any data. After finishing all pairwise comparisons, the user can review the results in the third tab.
For each node, existing in the decision model, the tool will generate a set of two plots: a bar chart indicating the utility levels of the two alternatives being compared and a pie chart showing the weights for each of the sub-categories. The last tab provides the user with the option of completing an additional cost-weighted analysis. This tab may be used only after all cost criteria have been eliminated from the decision model constructed using the first (left most) tab.
Pairwise comparisons are used to determine the relative importance of each criterion and the preference for each alternative when a set of criteria is considered. Each choice is a linguistic phrase. Some examples of linguistic phrases that can be used are: "A is more important than B", or "A is of the same importance as B", or "A is a little more important than B", and so on. In the pairwise comparison window, the user still has visibility to the decision hierarchy (in read-only mode) on the right hand side of the window. By clicking on each item in the hierarchy listing on the right, all pairwise comparisons associated with that level of the hierarchy will be displayed. Prior to viewing the results, the user must save the entire sets of comparisons by clicking on the "Save Comparison" button at the bottom of the window.
The user can compare criteria in two ways. For qualitative criteria, the user can use the scales provided on the form to rate the relative importance of criteria. If the criteria are quantitative (and accurate values or estimations are available) or the user wishes to use even scale numbers, text entry boxes are provided next to each comparison. The value entered in this box will represent the relative importance of the criterion on the left, over the criterion on the right. If the user uses both the radio buttons and manually enters a rating or a ratio in the text box, the value entered in the text box has the priority and will override the value entered using the scale.
From the results tab, the user can review the results of the AHP analysis completed using the pairwise comparisons or ratios entered by the user. At the center of the page, the overall preference for the two alternatives being compared is presented using percentages. The alternative with higher percentage value has the highest preference (i.e. utility).
These plots are dynamically generated for each hierarchy node. In other words, every time the user selects a node or category from the decision hierarchy on the right, the associated plots are drawn automatically. The bar chart represents the preference or utility level, calculated for the alternatives, by only considering the criteria within that specific node. In other words, when the user selects the "Schedule Constraints" node from the hierarchy, the stacked plot will show the utility level by only considering the weights and preferences associated with calendar and schedule constraints. By default, the software will display the plots for the overall results (plots associated with the "Goal" node) when the user moves on to the results tab. At the "Goal" node level, every generated bar is aggregated based on the utility values calculated using every subcategory. The pie chart shows the synthesized weight for every sub-criterion in a selected category. If the user selects a third-level criterion, the pie chart will show the user entered preference for different alternatives. The criterion with the highest weight in pie chart has the greatest contribution towards the total utility displayed in the stacked bar chart.
Toni L. Doolen, PhD
Oregon State University
Benjamin Tang, P.E.
Oregon DOT, Technical Services