Multiple Classification Analysis for Trip Production Models Using Household Data: Case Study of Patna, India

Abstract

Trip production models in India have traditionally been developed using simple regression analysis with population at census ward level as the independent variable. The current study developed multiple classification analysis (MCA) tables at the household unit level for trip production considering several other important variables that affect trip production. The city of Patna in India is taken as the case study, and its household data is considered for analysis. Households are further disaggregated into slum (low income) and nonslum households, and scenarios within them are considered for analysis. Nomograms are developed based on MCA tables and can be used to estimate trip rate values for other cities with similar socioeconomic characteristics. Slums and nonslum households revealed similar trip rate patterns, with the household size having the maximum impact on trips produced. Income and vehicle ownership show little effect on trip production rates. Also, MCA and linear regression models resulted in similar trip rates and accuracy. Hence, MCA is recommended to be adopted because it gives a more disaggregated output that is more stable over time and is easier to use because the values are readily available without further analysis.