Method and a device for customizing public transport line

Abstract

The invention discloses a method and a device for customizing a public transport line and relates to the field of line planning. The method comprises the steps of obtaining predicted line demand information, generating corresponding influence parameters according to the demand information, inputting the influence parameters into a trained passenger flow volume prediction neural network model to obtain a predicted passenger flow volume, obtaining a preset starting station, a preset ending station and the related predicted passenger flow volume, selecting a station conforming to the predicted passenger flow volume as a next station from the preset starting station until reaching the preset ending station, and forming a predicted line according to an optimization condition. The attendance rate of an unopened line is predicted through historical passenger flow data, therefore, a powerful basis is provided for customizing a new public transport line, meanwhile, the passenger flow demand ofan opened customized public transport line can be accurately predicted, and an important basis is provided for urban public transport managers to carry out overall public transport network layout planning and master the development direction of customized public transport.

Description

technical field [0001] The invention relates to the field of route planning, in particular to a method and device for customizing public transport routes. Background technique [0002] Public transportation passenger flow forecasting plays a very important role in the field of public transportation management and planning. For example, many public transportation route planning processes will use the forecasting of passenger flow. At present, there are many prediction methods for passenger flow. The commonly used forecasting models include time series methods , artificial neural network, support vector machine, decision tree, etc., its algorithms and models are relatively mature. However, most of the current research is based on long-term passenger flow analysis, and the number of studies on short-term public transport passenger flow forecasting is relatively small. With the intelligent development of public transport in my country, the demand for short-term public transport ...

AI Technical Summary

Problems solved by technology

However, most of the current research is based on long-term passenger flow analysis, and the number of studies on short-term public transport passenger flow forecasting is relatively small. With the intelligent development of public transport in my country, the demand for short-term public transport passenger flow forecast is increasing. However, due to the poor adaptability of existing model algorithms to deal with short-term special situations, the randomness of special situations is high, making it difficult to guarantee the accuracy and timeliness of the prediction model

At present, the passenger flow forecast of the designed lines at home and abroad often adopts linear weighted prediction of the passenger flow of the original bus OD line, but the generation of passenger flow is a relatively complicated process, and there are many factors that affect passenger flow. It is not enough to rely solely on the original bus OD flow

Images