Sharing bike attraction and generation prediction method based on ARIMA

Abstract

The present invention discloses a sharing bike attraction and generation prediction method based on an ARIMA (Autoregressive Integrated Moving Average Model). The method comprises the following stepsof: 1) collecting GPS positioning data of static parking positions of available bikes in an area, and continuously performing collection for assigned days; 2) obtaining geographic information data oftraffic zones in the area; 3) matching geographical location information of the sharing bikes to each traffic zone; 4) establishing a sharing bike trip total sample; 5) establishing a sharing bike available bike distribution spatial and temporal distribution thermodynamic chart, and a space thermodynamic chart of the number of times of attraction and generation of each zone; 6) establishing a timesequence of the number of travel times of each zone; 7) establishing an ARIMA prediction model after parameters are calibrated; and 8) predicting a sharing bike trip of each traffic zone in a next time aggregation interval. The demand prediction method employs position data of bikes to sense time-space features of the sharing bikes in a city and performs prediction of the time sequence so that adata support is provided for operation, management and scheduling of the sharing bikes.

Description

technical field [0001] The invention relates to an intelligent management method for shared bicycles, in particular to a method for predicting short-term attraction and occurrence of shared bicycles based on an ARIMA model. Background technique [0002] Since the end of 2016, dockless shared bicycles began to appear in the streets and alleys of major cities. You only need to scan the QR code to pay the fare, and you can use and park it anytime and anywhere. Shared bicycles have profoundly changed the city's travel structure. The return of bicycles to cities has brought convenience to urban residents' travel, but it has also brought difficulties to the operation and management of government management departments and shared bicycle rental companies. For example, during peak hours, due to the tidal nature of travel, the supply of shared bicycles in popular travel areas will be in short supply, and it is difficult to find a car, while the number of shared bicycles in popular ar...

AI Technical Summary

Problems solved by technology

At present, there is no mining and forecasting method for the occurrence and attraction of natural communities for the massive data of shared bicycles.

Images