working paper

Neural Network Models For Automated Detection Of Non-recurring Congestion

Publication Date

December 31, 1992

Author(s)

Abstract

This research addressed the first year of a proposed multi-year research effort that would investigate, assess, and develop neural network models from the field of artificial intelligence for automated detection of non- recurring congestion in integrated freeway and signalized surface street networks. In this research, spatial and temporal traffic patterns are recognized and classified by an artificial neural network.