作者: Brian L. Smith , R. Keith Oswald
关键词:
摘要: This paper explores the ability of imprecise computations to address real-time computational requirements in infrastructure control and management systems. The research this area focuses on development nonparametric regression as a means forecast traffic flow rates for transportation Nonparametric is forecasting technique based nearest neighbor searching, which forecasts are derived from past observations that similar current conditions. A key concern regarding significant time required search neighbors large databases. Results presented herein indicate approximate neighbors, applied regression, may be used adequately speed execution with acceptable degradations accuracy.