Research on Personalized O2O Instant Logistics Time Prediction Model based on Time Series Analysis and Spatial Continuity

In order to make better use of spatio-temporal correlation in the model, this paper proposes a spatio-temporal transportation vehicle driving state matrix model VDSM. At the same time, a personalized O2O instant logistics time prediction model based on time series analysis and spatial continuity was proposed. Finally, in the experimental part, 10 transportation lines of trunk and non-trunk sections are selected, and the logistics in-transit time prediction model based on VDSM is used to predict the logistics in-transit time on the historical data sets of these lines. The final experimental results show that the strategy adopted by the personalized O2O instant logistics time prediction model based on time series analysis and spatial continuity can not only effectively reduce the overall prediction error of the model, but also increase the stability of the model prediction.