Based on BP neural network model and queuing theory to build taxi driver decision model

Abstract

In this paper, the decision-making model based on BP neural network and queuing theory is used to solve the decision-making problem of taxi driver whether to go to the airport to carry passengers. Using BP neural network model, this paper analyzes the influence of season, time, urban GDP, urban population, urban consumption level and other factors on the locomotive occupancy rate; analyzes the influence of airport location, weather, alternative means of transportation and other factors on the taxi selection rate, and obtains the demand of taxi based on the probability of different passenger pairs; compares the "storage pool" taxi The number of cars, according to the taxi driver's minute salary level and working time, is predicted by queuing theory in the period of joining the "car storage pool waiting" and seeing off passengers, and the decision is made by comparing the expected income in the same period of time.