Identifying Traffic Congestion Patterns of Urban Road Network Based on Traffic Performance Index

Urban congestion has become a global problem with urbanization and motorization. The analysis of time-varying traffic patterns is necessary to formulate effective management strategies. existing studies have focused on flow developed by the volume, speed density road sections in limited district, while long-time macro network at city level inadequate. This paper aims recognize urban based performance index (TPI) 699 days 2018, 2019 2021 Beijing. self-organizing maps (SOM) method improved an automatic clustering number determination algorithm proposed cluster TPI. clustered into Mondays, Fridays, ordinary weekdays, congested weekdays winter summer vacation, Saturdays, Sundays festivals patterns. Patterns Mondays prominent morning peak, vacation evening peak. are less than weekday It verified that SOM this clusters more detailed accurate patterns, it applicable TPI different years. degree increases 7.15% peak hours decreases 7.50% off-peak compared due COVID-19. helpful for terms making decisions according