Technologies for bridge structure design-based knowledge reuse

Managing and reusing knowledge in bridge structure designs leads to greater efficiencies and improved designs.However,this knowledge has not been effectively reused due to a lack of ways to capture the design knowledge.This paper develops a knowledge reuse framework after analyzing bridge databased object oriented methods.The framework provides three layer similarity functions for bridge projects,structures,and elements using the K nearest neighbor(KNN) method for similar bridge designs.A case modification technique was developed based on neural networks and multivariate regression for new design requests.Finally,tools were developed for automating drawing arrangements and for manual arrangements.