Combined forecasting of regional logistics demand optimized by genetic algorithm. Issue 2 (29th July 2014)

• Record Type: Journal Article
• Authors: Naiming Xie, Dr Yingjie Yang and Dr Chuanmin Mi, Professor; He, Feng-biao; Chang, Jun
• Is Part Of: Grey systems. Volume 4:Issue 2(2014)
• Page Start: 221
• Page End: 231
• Publication Date: 2014-07-29
• Subjects: Cybernetics -- Periodicals; Systems engineering -- Periodicals; 003.5
• DOI: 10.1108/GS-04-2014-0011
• Languages: English
• ISSNs: 2043-9377

Abstract

<abstract> <title> <x content-type="archive" xml:space="preserve">Abstract</x> </title> <sec> <title content-type="abstract-heading">Purpose</title> <p> – The purpose of this paper is to establish a combined forecasting model to predict regional logistics demand, which is an important procedure on decision making of regional logistics planning. </p> </sec> <sec> <title content-type="abstract-heading">Design/methodology/approach</title> <p> – There are several kinds of mathematical models often used in forecasting regional logistics demand. Trend extrapolation method extrapolates the future development trends bases on the hypothesis that the regional logistics will develop steadily. Grey system method predicts the change of logistics demand by the generation and development of original data sequence and excavation of inherent rules of the original data. Regression method obtains the change rules through the analysis between explained variable and explanatory variables. Each method has unique characteristics. In order to improve the accuracy of the prediction, combined methods are established. Genetic algorithm is used to determine the weights of different single models. </p> </sec> <sec> <title content-type="abstract-heading">Findings</title> <p> – The results show that the combined forecasting model optimised by genetic algorithm can improve the accuracy. </p> </sec> <sec> <title content-type="abstract-heading">Practical implications</title> <p> – Combined forecasting model can integrate the advantages of different single forecasting models. The key of improving the accuracy is to determine the weights of single forecasting models. Genetic algorithm can do well in finding suitable weights of each single forecasting model. </p> </sec> <sec> <title content-type="abstract-heading">Originality/value</title> <p> – The paper succeeds in providing a combined forecasting model using genetic algorithm to determine the weights of each single prediction model, which helps to the decision making of regional logistics demand.</p> </sec> </abstract>