This study analyzes commuting patterns in urban rail transit and investigates the characteristics and regularities of different commuter types. The analysis is based on Changsha Metro AFC data, covering two consecutive weeks of working days from March 21 to April 3, 2023. A Gaussian Mixture Model (GMM) is constructed using three variables: morning boarding time, evening boarding time, and the average commuting duration. This model classifies passengers into three categories: the classic commuting pattern, the offpeak commuting pattern, and the longdistance commuting pattern. The results show that classic commuters primarily commute during regular peak hours (7:30–8:30 and 17:30–18:30), while longdistance commuters tend to start slightly earlier, and offpeak commuters avoid peak periods, with their morning boarding times spanning from 7:00 to 12:00 and evening boarding times concentrated between 20:00 and 22:00. In terms of commuting duration, classic and offpeak commuters typically travel for 15 to 30 minutes, whereas longdistance commuters predominantly travel for over 30 minutes. Additionally, the residential and workplace station distributions highlight that longdistance commuters are more likely to reside in peripheral urban areas, and the workplace stations of offpeak commuters are more concentrated than those of the other two groups, predominantly located in areas bustling with dining and entertainment activities.