01024487

Component

The methodology to model freight transportation implications on communities is always an important issue in transportation planning evaluation. This paper develops a neuro-fuzzy logic modeling method to assess the freight transportation implications on communities. An expert opinion survey and an agency data survey were conducted to obtain the knowledge rule base and input/output data, which made the modeling stay in a balanced position between both quantitative and qualitative analysis. Sugeno Fuzzy Models were built within MATLAB programming platform using Fuzzy Logic Toolbox, and ANFIS technique was also applied to calibrate the membership function parameters through data training and testing. The performance index (PI) was derived from the implication assessment systems. The conclusion is that the neuro-fuzzy logic based modeling is a good methodology for the analysis of freight transportation implications on communities. While only freight-related factors are considered in this paper, non-freight-related factors may also be incorporated for better effects as there is no limitation on the input data types for the model. Such a methodology could be applied to many other impact analyses (e.g., economy implications on traveler behaviors, etc.).

01020180

06-1291

English

26p

2006

Transportation Research Board 85th Annual Meeting

CD-ROM

Figures; References; Tables

Freight transportation; Input output models; Performance evaluations; Performance measurement

Sugeno Fuzzy Model

Community impact assessment; Freight demand modeling; Neuro-fuzzy networks

Economics; Freight Transportation; Highways; Planning and Forecasting; Railroads; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2006 Paper #06-1291

TRIS, TRB

Mar 3 2006 10:38AM