Airline ticket seat reservation number prediction method based on improved increment model

Abstract

The invention relates to an airline ticket seat reservation number prediction method based on an improved increment model, which comprises the steps of: (1) acquiring and preprocessing flight historical seat reservation data; (2) acquiring cyclic period data and corresponding period data; (3) predicting an airline ticket seat reservation number. According to the invention, corresponding freight space weighted prediction results are obtained by utilizing cyclic period dates and corresponding period dates, and a final seat reservation number prediction result is obtained by combining a multiplelinear regression model. Compared to the prior art, according to the invention, a freight space number weighting method is introduced, and a problem of interference of different total flight freight space numbers on the prediction result, which is caused by different flight numbers and different flight types on different dates, is solved. Moreover, connection between corresponding period and cyclic period seat reservation data and a to-be-predicted seat reservation number is sufficiently utilized, inherent time characteristics of historical data are mined, prediction accuracy is improved, andthe airline ticket seat reservation number prediction method has the important effect of guiding an airline company to carry out subsequent freight space regulation control and dynamic pricing and hasthe important practical application significance.

Description

technical field [0001] The invention relates to the technical field of civil aviation information, in particular to a method for predicting the number of seat reservations of airline tickets based on an improved incremental model. Background technique [0002] The core of revenue management is to sell the product to the right customer at the right time at the right time in order to obtain the greatest economic benefits. Aviation revenue management means that airlines use demand forecasting and dynamic pricing to sell every seat on a flight at the most appropriate price, thereby maximizing revenue. Airlines make demand forecasts based on historical flight reservation data, formulate corresponding air ticket prices on this basis, and balance supply and demand through strategies such as stock control and overbooking. Demand forecasting is the basis of aviation revenue management, and airlines cannot implement scientific dynamic pricing, seat allocation and other decisions with...

AI Technical Summary

Problems solved by technology

[0005] Although the traditional incremental forecasting model is simple and easy to implement, it has the following disadvantages: it simply uses all the existing historical data for forecasting, treats all historical data equally, and does not consider the timeliness of historical data, such as ring period, Important features of contemporaneous data

Short forecast time horizon and inaccurate forecast results

Images