Post Epidemic Traffic Vitality and Economic Recovery Based on DTW and K-medoids Clustering Method

XU Le; MA Jianxiao; LI Gen; XU Chubo; BAI Yingjia

doi:10.3969/j.issn.1672-9722.2023.10.014

Abstract：After the COVID-19 outbreak,cities had different recovering modes.The cities'vitality recovering index is used to begin the study,the purpose of this paper is to explore the problem appeared during the period of recovering.Firstly,python is used to run DTW(Dynamic Time Warping)and K-medoids to cluster 30 cities'vitality index,the character of each modes is tried to found out,then cities'level、population、cumulative infection numbers etc.are used to explore the similarity of the cities under the specific mode.By comparing the urban economic indicators in the peak period of the epidemic(the first quarter of 2020 statistical in-dicators)and the urban economic indicators in the epidemic impact and recovery period(the first half of 2020 statistical indica-tors),this paper explores the problems in the process of urban economic recovery,and finds that:1)there are three recovery modes in 30 cities,and each recovery mode has its own characteristics,Mode 0 fluctuates greatly and lasts for a long time,Mode 1 fluctuates in the early stage and becomes stable in the later stage,Mode 2 is relatively stable.2)Under the same recovery model,there is a strong correlation between the level of cities and their relationship,while the scale of cities and the cumulative number of infections have an impact on the volatility of urban recovery.3)In the process of urban recovery,there are significant differences in production and price levels in different periods.Based on the above conclusions,this paper provides policy suggestions for the city to resume production and life when encountering the epidemic in the future.