Answers are grounded in this paper and cite their sources.
This paper presents a methodology for deriving representative motorcycle driving cycles in urban areas with complex topographies, where variables such as changes in slope or altitude above sea level impact mobility behavior, using computational intelligence and geospatial data. Focusing on the Aburrá Valley (Medellín, Colombia), the research incorporates data from 122 trips and 27 riders across various motorcycle models, weekdays, and times of day. The statistical analysis included selecting 1863 micro-trips with the most representative routes derived from heat maps, and examining mobility parameters such as speed, positive and negative accelerations, idle time, and moving time. Advanced machine learning techniques, including Support Vector Machines (SVM), Principal Component Analysis (PCA), and the LAMDA fuzzy supervised classifier. The results demonstrated a separation of the data into four well-defined groups, showing high sensitivity and specificity, resulting in high precision and accuracy (94.37 %). The methodology yielded two driving cycles reflecting urban (CMVA) and mixed urban-motorway (CMVA+M) conditions. This approach supports the development of transport policies and sustainable mobility strategies by integrating artificial intelligence and spatial data analysis to enhance understanding of two-wheeled transport dynamics in urban environments shaped by variable terrain.
No takes yet. Share an insight, caveat, or question.
Londoño et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: