Exploring the travel behavior of household-owned electric bikes with the consideration of nonlinear effects of factors with the aid of the generalized additive mixed model

Abstract

Electric bikes (E-bikes) reduce traffic burden in developing countries and provide a lightweight travel option for households. Few studies examined nonlinear effects of factors on e-bike travel behavior. This study adopts the generalized additive mixed model to explore the travel behavior of household-owned e-bikes with the consideration of nonlinear effects of factors based on data from Zhoukou, China. The study confirms that there is a significant nonlinear effect of many factors on the travel behavior of household-owned e-bikes, with the nonlinear effects varying by factors. The lower the number of household bicycles, the higher the travel frequency of household-owned e-bikes. An increase in the number of household cars leads to a decrease in the travel frequency of household-owned e-bikes, and this effect becomes larger as the number of cars increases. The number of household-owned e-bikes is positively correlated with travel frequency, but the correlation weakens after more than 3 e-bikes. There is a negative correlation between the travel frequency of household-owned e-bikes and the number of persons older than 6 when it exceeds 4. The effect of the number of persons older than 6 on the travel frequency of household-owned e-bikes gradually increases when the number of persons older than 6 exceeds 3. Finally, this study puts forward policy suggestions regarding enhancing the attractiveness and competitiveness of household-owned e-bikes. These findings could advance the understanding of the role of e-bikes in households and serve as references for policymakers and planners.