In 2020, the outbreak of COVID-19 pneumonia has a great impact on China's economic and social life. Construction and transportation industry has been greatly impacted and suffered from its high mobility. With the trend of the epidemic situation getting better, it is necessary to study the accuracy and efficiency of the transportation flow prediction method after the epidemic to reasonably adjust the transportation capacity and allocate resources, so as to reduce the economic losses and make contributions to the epidemic prevention and control. This paper studies the big data of migration between Xi'an and Chengdu from January 1, 2020 to March 15, 2020, and divides the epidemic situation into four stages according to the introduction of the elastic coefficient according to the development of the epidemic situation. In each stage, the elastic coefficient of index change is introduced in combination with the decreasing impact of epidemic prevention and control measures on transportation flow. Finally, a modified moving average method is formed, which is compared with the ordinary moving average method by using the Hadoop big data platform, using MapReduce programming model to run the modified moving average method and running time under the condition of a single computer, the prediction efficiency is tested by using the acceleration ratio. The results show that the modified moving average method combined with Hadoop big data platform can improve the accuracy and efficiency of the intercity transportation flow prediction under the epidemic situation.

CICTP