作者: A.S. Fowkes , A.L. Bristow , P.W. Bonsall , A.D. May
DOI: 10.1016/S0965-8564(97)00029-3
关键词:
摘要: This paper describes a methodology which permits optimal strategies for strategic transport models to be found by use of limited number model runs together with regression modelling the resulting response surface. Typically, it will case that policy variables is sufficiently large cannot run all possible combinations their levels. Furthermore can very difficult interpret results from where there are lot changing levels between runs. The proposed surface specifically in locality optimum, thereby greatly clarifying what should further tested model. A study, city Edinburgh, indicates this identify improved compared conventional methods, even when used far fewer than methods.