The Impact of the Modifiable Area Unit Problem on Online Ride-hailing Travel Behavior Analysis

As the built environment gradually deepens the analysis of online ride-hailing travel behavior, it is necessary to further explore the impact of modifiable areal unit problem (MAUP) on behavior analysis modeling. Therefore, this article selects Didi travel data and multi-source spatial data in Haikou City to construct built environment and socioeconomic attribute indicators, and analyzes the impact of the built environment on morning and evening peak passenger travel by constructing a geographically weighted regression (GWR) model. On this basis, the impact of MAUP on behavioral analysis is explored by comparing different modeling results. The results show that Moran's I show a fluctuating trend as the scale increases; the interpretability and goodness of fit of the model gradually improve as the scale becomes larger; from the comparative analysis of the regression coefficients of GWR, it can be concluded that variables such as Transport Facilities, Accommodation Hotel, and Healthcare are greatly affected by spatial scale, and the impact on these variables should be considered in modeling analysis. This study can provide a theoretical basis for the analysis of online car-hailing travel behavior and better explore the impact of the built environment on online car-hailing travel.