Application of knowledge-based expert systems for rating highway bridges

This paper presents the establishment of a Bridge Rating Expert System (BRES) by using a knowledge-based computer artificial intelligent expert system for analysis and rating of existing short span prestressed concrete highway bridges. The rating methods considered were: Inventory Rating, Operation Rating, Rating Factor Rating and Sufficiency Rating. A strength rating is based on the evaluation of existing prestressed concrete bridges in accordance with the American Association of State Highway Transportation Officials (AASHTO) specifications, and inventory rating for all bridges according to the Federal Highway Administration (FHWA) guide for the Structure Inventory and Appraisal of the Nation's Bridges. The parameters considered were bridge type, dead loads, type of vehicle loading, bridge dimensions and cross-section, material properties, reinforcement details, and existing physical condition of the bridge. The results show that, through its runtime rule controller, user interface and linkage mechanisms, the BRES is a friendly user system for rating highway bridges. In addition, the EXSYSP shell provides an effective tool for other possible applications in civil engineering practice.