THE CONCRETE BRIDGE MAGAZINE

Winter 2019

ASPIRE is a quarterly magazine published by PCI in cooperation with the associations of the National Concrete Bridge Council. The editorial content focuses on the latest technology and key issues in the Concrete Bridge Industry.

Issue link: http://www.aspiremagazinebyengineers.com/i/1064399

Contents of this Issue

Navigation

Page 31 of 51

F H WA 30 | ASPIRE Winter 2019 Many of our nation's bridges have aging decks, and bridge owners are struggling to keep these decks in satisfactory condition. Bridge decks di- rectly support vehicular and truck traffic and have large surface areas that are exposed to the environ- ment, which makes them among the most diffi- cult bridge components to maintain. Bridge own- ers are constantly searching for ways to build and maintain long-lasting decks. e research project described in this article used the National Bridge Inventory (NBI) database to study concrete bridge deck performance and develop probabilistic ser- vice-life prediction tools which may assist owners in maintaining their concrete bridge decks. Nationwide Concrete Bridge Deck Performance Inventory e NBI database includes data on every highway bridge on public roads in the United States (615,002 bridges in the 2017 edition). After filtering this massive database, researchers found 150,136 suitable concrete bridge deck ser - vice-life data strings between the years of 1992 and 2014. ere are 116 data fields for each bridge in the NBI. e researchers identified 21 param - eters and computed four others that offer useful service-life data. ese 25 parameters were com- piled for each of the 150,136 service-life data strings to develop a database that the researchers titled the Nationwide Concrete Highway Bridge Deck Performance Inventory (NCBDPI). Table 1 lists 18 parameters; 14 that came from the NBI and are included in the NCBDPI, and the four others were computed from NBI data. Time-in-Condition Rating e bridge deck condition rating (CR), NBI item 58, provides a numerical assessment of the bridge deck's condition on a scale from 0 (failed condition) to 9 (excellent condition). is assess- ment is typically made by visual inspections in which inspectors use indicators such as cracking, spalling, leaching, delamination, and full or par- tial-depth failures to assign a CR. e researchers identified a string of unchanged condition rat- ing assessments occurring over multiple years as a time-in-condition rating (TICR). erefore, TICR is defined as the duration of time in years that a CR remains constant before it decreases, indicating further deterioration. Sample CRs for three hypothetical concrete bridge decks for the period from 1992 to 2014 are shown in the figure on the opposite page. A TICR can be visually iden- tified in the figure as any plateau in the data string. e ideal service-life TICR data string falls com- pletely within the time boundaries of the research period (1992–2014) and shows no anomalies such as gaps or sudden drops and rebounds in the CR value. A significant proportion of TICR data (ap- proximately two-thirds of the CR from the NCB- DPI database) is "censored," meaning that the plateau forming the TICR is only partially observ- able. A key attribute of this research was the proper treatment of censored data. Examples of censored and uncensored TICR strings for the three hy- pothetical bridge decks are shown in the figure. Discarding censored data or treating them as un- censored data would introduce significant bias. In contrast, the approach used by the researchers re- sulted in a more robust and accurate analysis. Bridge Deck Survival Analysis Survival analysis is routinely used in other in- dustries to analyze data in which the time until an "event of interest," such as death, divorce, or fail- ure of a mechanical part, occurs. In this research, the "event of interest" is a drop in the CR. In this study, a Bayesian survival analysis was used to compute the probability that the CR of a bridge deck with specific characteristics remains un- changed. e analysis results were used to generate survival curves for varying parameters. ese sur- vival curves provide practical data in an illustrative graph. e research found that varying the param- eters of climatic region, average daily truck traffic (ADTT), maintenance responsibility, and struc- ture type generated the most noticeable changes in TICR survival. An example of hypothetical concrete bridge deck survival curves is shown in a figure on this page. is curve was generated Concrete Bridge Deck Service-Life Prediction Tools by Reggie Holt, Federal Highway Administration Table 1. Nationwide Concrete Highway Bridge Deck Performance Inventory (NCBDPI) Parameters and Data Sources NCBDPI Parameter Data Source or Corresponding National Bridge Inventory Location County (parish) code Item 3 Structure number Item 8 Maintenance responsibility Item 21 Functional classification of inventory route Item 26 Lanes on structure Item 28 Structural material/design Item 43a Type of design and/or construction Item 43b Deck condition rating (CR) Item 58 Designated inspection frequency Item 91 Deck structure type Item 107 Type of wearing surface Item 108a Type of membrane Item 108b Deck protection Item 108c Average daily truck traffic (ADTT) Item 109 Deck area Item 49 × item 51 Climatic region Assigned using the IECC Distance to seawater Calculated based on coastline and bridge location items 16 and 17 Bridge age Based on items 27 and 106 Note: IECC, International Energy Conservation Code . Table: Federal Highway Administration.

Articles in this issue

Archives of this issue

view archives of THE CONCRETE BRIDGE MAGAZINE - Winter 2019