Keywords: maintenance, highway, bridge, expert system, management, decision making.

The aim of this work is to develop a computer system that can represent human expertise regarding maintenance decisions in highway bridges. Such a system can provide decision support to maintenance agencies in the management process. In addition, the procedure to knowledge acquisition and coding can reveal the importance of each parameter involved in the decision making process towards further development of a highway management system. Previous research efforts reported in the literature have led to the development of a few expert systems for bridge management [1,2,3]. A limitation of such systems is that they typically deal with specific parts of the bridge management process rather than with the whole process.

In this work an expert system is presented which incorporates several parameters involved in the decision making process such as distress type, severity and extent, road functional classification and traffic loads, bridge structural characteristics (superstructure type, foundation type, river bed characteristics), bridge age, and environmental conditions. The proposed system consists of three modules, one for setting maintenance priorities, another for determining of alternative applicable treatments, and a third for selecting the appropriate treatment.

The first module records all parameters which are considered in the decision making process along with their relative weights. Each bridge is considered with respect to every parameter and a value is assigned depending on how the bridge is rated with regard to this parameter. Following a multi-criteria type of analysis, a total index is computed for each bridge which indicates the maintenance priority. Parameter weights and bridge rating are at this stage set by experience and checked by a trial and error technique. The objective of such operation is to reach an acceptable level of coincidence regarding priority ranking between the system output and the actual expert decisions.

The second module investigates all applicable treatments considering the specific conditions of each bridge. A rule-based system has been developed to represent the expert's decision making under each possible condition. The system employs forward reasoning as an inference engine and may be depicted in a form of a decision tree. Besides identifying applicable treatments, the system also sorts these treatments in each case according to their effectiveness/cost ratio. Effectiveness is considered in terms of the expected lifetime of each treatment until another treatment will be necessary.

The resource allocation module aims to select the most appropriate maintenance strategy for a number of bridges based on the output of the previous modules and considering any budget constraints. In particular, starting from the bridge with the highest maintenance priority, the best cost-effective treatment among all applicable ones is proposed. The process continues until the available budget is allocated. If the number of bridges that are treated is less than desirable, other less costly treatments may be selected so that more bridges are maintained.

The system has been evaluated with actual and simulated data (generated through a Monte Carlo technique). Test results indicate that it can provide a valuable tool for short-term maintenance decisions as it provides a basis for consistent decisions and can handle a huge number of bridges.

1

Zuk, W., "Expert system for determining the disposition of older bridges", Transportation Research Record, 1290, 145-148, 1991.

2

Brito, J., Branco, F.A., Thoft-Christensen, P. and Sørensen, J.D., "An expert system for concrete bridge management", Engineering Structures, 19(7), 519-526, 1997. doi:10.1016/S0141-0296(96)00125-3

3

Issa, M.A, Tsui, S. and Yousif, A., "Application of knowledge-based expert systems for rating highway bridges", Engineering Fracture Mechanics, 50(5/6), 923-934, 1995. doi:10.1016/0013-7944(94)E0067-Q

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £82 +P&P)