<p>With the accelerated urbanization process, accurately understanding urban travel demand has become increasingly important. This study proposes a Hotspot Detection-based Analysis Method (HDAM) for key drivers of travel demand, which aims to minimize the impact of the Modifiable Areal Unit Problem (MAUP) on travel analysis. Focusing on urban multi-density distributed data, HDAM effectively identifies local travel hotspot areas through an adaptive hotspot detection method and constructs buffer zones around these hotspots as analysis units. On this basis, the study further proposes a hierarchical modeling approach to explore the key driving forces influencing travel demand under different levels, forming a data-driven framework for analyzing influencing factors. The results demonstrate that the HDAM method can effectively capture the spatial distribution characteristics of travel demand, and the hierarchical modeling approach significantly enhances the accuracy and reliability of the analysis. The method proposed in this paper helps to deeply explore the formation mechanism of travel hotspots and provides a scientific basis for urban planning and traffic management.</p>