Keywords: pavement, maintenance, management, expert system, knowledge-based system, decision making.

Pavement management systems (PMS) have been widely used in recent decades to evaluate pavement conditions over time, determine alternative maintenance and rehabilitation strategies for given roadway sections, and to recommend "best" or "most cost effective" network-wide strategies over a planning horizon. Considering the complexity of the typical pavement management problem and the difficulty to establish realistic constituent models and parameter assessment, the experience of maintenance professionals may be invaluable in the development of new or the enhancement of existing algorithmic PMSs. Existing research on knowledge-based pavement management systems has primarily been directed to highway and arterial road networks. Urban roads, however, present some special characteristics that may require particular attention. Among them, pavement design and construction standards, traffic composition (e.g. increased bus proportion), pavement loading patterns (e.g. frequent stop-go traffic), and frequent utility network cuts that wound the pavement integrity can be mentioned.

The aim of the present study is to develop an expert system which can assist engineering decision-making with regard to urban road pavement deterioration and provide proposals for cost-effective maintenance and rehabilitation strategies. The proposed system considers a number of parameters including pavement condition characteristics (e.g. distress type, severity, and extent), pavement age, soil type, traffic loads, environmental conditions, as well as treatment characteristics (e.g. treatment type, distress-treatment suitability matrix, treatment effectiveness and cost).

The proposed system consists of three decision modules. The first is to provide maintenance priorities among road segments. Here current pavement conditions, the traffic loads and other operating characteristics are considered. A weighted, utility-type of function is employed to indicate the maintenance priority. Parameter weights and section grading were set by experience and evaluated by experts on the basis of results from the actual data entries. The second module explores all applicable treatments in each section considering the specific conditions in it. A rule-based expert system has been developed to represent the expert's opinion under each possible condition. The system employs forward reasoning as an inference engine and consists of 6,480 "if-then" rules. The system further sorts these treatments in each case according to their effectiveness to cost ratio. The third module calculates the cost of the necessary maintenance works and provides a network-wide maintenance and rehabilitation plan taking into account any budgetary constraints.

The system has been evaluated with data from the road network of the City of Patras. Results have been appraised by expert engineers and indicate that the proposed expert system can provide a reliable tool for short-term maintenance and management decisions.

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 £50 +P&P)