Identifying bridge structural condition development trends via categorical inspection condition rating with case studies

Abstract

Inspection data in a bridge management system (BMS) may provide us with comprehensive information about a structure's condition history. Identification of bridge structural condition development trends from historic inspection data for maintenance planning has drawn more and more attention, particularly from bridge agencies, who prefer ‘worst-first’ maintenance and rehabilitation strategies. In this paper, rather than correlating inspection data with fundamental structural deterioration models, a statistical inspection data-based analysis was developed. To demonstrate the inspection data-based approach in identifying bridge structural condition development trends in practical scenarios, case studies of concrete deck slabs and steel beam girders were applied. The ANalysis Of VAriance (ANOVA) technique was adopted to identify the major factors considered by a BMS to have a significant influence on the condition index. As a result, three major factors, bridge construction year, inspection year and inspector, were identified when structures were exposed to the same environment. Associated with these major factors, statistical indicators were introduced to characterize the condition development trends of structures, and to detect signs related to the potential of threat to the structural integrity. The indicators were further applied to identify a specific subgroup, e.g. those bridges that were constructed in a certain period and experienced the most deterioration or the highest deterioration rate. The approach developed can be extended to look into more causal factors on how they may alter the structural condition development trends when their information is available within inspection records. The outcomes may offer useful information for maintenance planning. Furthermore, it was found that the inspector also had quite a considerable impact on the condition index. A difference in structural condition assessment among inspectors may occur regardless of the same structural condition scenarios. This difference was quantified and applied to identify inspectors who tended to underestimate deterioration in structural condition.