Alternative driver order forecasting method and agent driver capacity scheduling method

Abstract

The invention discloses a driving order prediction method and a driving capacity scheduling method, which can be used in airports. The driving order prediction method includes: establishing a historical order database for each airport; performing regional clustering of airports based on order change similarity; classifying each regional category According to the order volume in the basic time period, multiple basic time periods are clustered by time period, so that different order prediction base time periods are formed under each regional class; for each order prediction base time period of each regional class, Extract the corresponding order volume and change factors, and use BP neural network for nonlinear fitting, so that each order prediction reference time period of each regional class obtains its own different order prediction model; when receiving an order prediction request, according to the requested Based on the regional category and order forecasting base time period, select the corresponding order forecasting model and predict the order volume by combining the change factors. The driving capacity scheduling method generates a reasonable driver allocation plan based on the above order prediction results.

Description

technical field [0001] The invention relates to the technical field of intelligent equipment for substitute driving, in particular to a method for predicting orders for substitute driving and a method for dispatching capacity of substitute driving based on data mining technology. Background technique [0002] Nowadays, more and more people choose to travel by car to the airport by plane. However, when parking at the airport by car, they will encounter the following problems: [0003] 1. Waste of time: Compared with people who take subways, taxis and other means of transportation, it takes time for the parking process, especially when the airport has a large passenger flow and the parking lot is often saturated, resulting in unexpected delays in the parking process precious time, or even miss the plane. When returning from the return journey, the parking position of the vehicle may be forgotten, which will bring a lot of trouble to the self-driving car owner; [0004] 2. Ex...

AI Technical Summary

Problems solved by technology

[0010] The main purpose of the present invention is to propose a method for predicting orders for surrogate drivers. Based on data mining technology, multi-dimensional analysis of historical orders is carried out to predict orders more accurately, so as to solve the problem of insufficient estimation and analysis of orders in the existing technology. The unreasonable technical problems caused by the allocation of substitute drivers

Images