1 Performance Evaluation of Concrete Slabs of Existing Bridges using Neural Networks Kei Kawamura a , Ayaho Miyamoto a , Dan M. Frangopol b, * , and Ryuichi Kimura c a Department of Computer & Systems Engineering, Yamaguchi University, Tokiwadai 2-16-1, Ube, Yamaguchi, 755-8611, Japan b Department of Civil, Environmental, and Architectural Engineering, University of Colorado, CB 428, Boulder, CO 80309 – 0428, USA c Kurimoto, Ltd. Ohamanishimachi 2-2,Sakai-Shi, Osaka 590-0977, Japan * Corresponding author. Tel: +1-303-492-7165; fax: +1-303-492-7317. E-mail address: Dan.Frangopol@Colorado.EDU (D. M. Frangopol) Abstract This paper presents a novel approach for developing a performance evaluation system for concrete slabs of existing bridges. The system evaluates the performance of concrete slabs under deterioration on the basis of expert knowledge. Characteristic features of this study are the definition of bridge performance, the performance evaluation system, and the use of neural networks. The proposed approach performs inference in the network, facilitates refinement of the knowledge base embedded in the system by the back propagation method, and prevents not only the inference mechanism of the system but also knowledge base after machine learning from becoming a black box. The numerical examples and conclusions reveal that the proposed approach demonstrates real potential for practical applications. Keywords: performance evaluation; durability; load-carrying capability; expert system; fuzzy; machine learning; neural network