Improving the Accuracy of the GM(1,1) by Data Grouping Technique and Its Application to Forecast Vehicle Volume and CO2 Emission in Tokushima City, Japan

Abstract

This paper deals with the problem for improving the accuracy of the grey model (GM(1,1)) in traffic flow and CO2 emission prediction. In order to improve the prediction accuracy, we adopt a data grouping technique along with the GM(1,1) and a Grouped GM(1,1) (GGM(1,1)) is established. Moreover, by applying techniques of accumulated generating operation (AGO) and inverse accumulated generating operation (IAGO) on training data collected from national route 11 of Tokushima City, Japan, the accuracy of GM(1,1) and GGM(1,1) in forecasting vehicle volume and CO2 emissions is investigated. Therefore, in this paper we contribute to develop and enhance the GM(1,1)’s accuracy in prediction.