Many urban areas contain a mix of ethnic groups or households with structural differences in income and social status. Current trip generation techniques do not explicitly allow for the treatment of these structurally different socioeconomic groups. A statistical approach that can be used in such cases is described. The proposed framework allows for the treatment of different ethnic household groups;each group can have different subgroups based on one or more qualitative features, such as dwelling unit type or income. A case study based on data from the kuwait metropolitan area is presented. Households are classified according to three nationality groups and four house types. The house type is found to be significant for only one ofthe groups. Differences in important quantitative household characteristics, such as car ownership and number of adults and children, appear both between and within subgroups. This paper appears in transportation research record no. 1328, Travel demand forecasting: new methodologies and travel behavior research 1991
Abstract