Zhejiang Province, Inbound Tourism Market Demand Study

Since 1990's, the inbound tourism has been developed rapidly in the province of Zhejiang. In 2002, the number of tourists to Zhejiang from abroad had increased to 2 million for the first time, which means the inbound tourism in Zhejiang had stepped into a new stage. With Beijing holding the Olympic game and Shanghai biding for the International Fair successfully, the city will be opened to the outside world further and will get more chances to propagate tourism and show itself. It's certain that the international tourism will continue to develop in Zhejiang day by day. As Inbound tourism can earn foreign exchange, and solve the problem of employment, it plays an influential role throughout the development of the tourism industry. It has an important guiding significance to research the dynamic trends of inbound tourism market in Zhejiang.What the thesis researches into is the inbound tourist source market of Zhejiang. According to some statistic information and materials obtained from some spot checks on inbound tourists, it analyses the general characteristic of tourist source market of Zhejiang, and makes cluster analysis of inbound tourist source market. Based on it, it makes a forecasting model for tourism demand of Zhejiang.How to establish a forecasting model for tourism demand is an important project of tourism study. Many researchers are all realizing that it is very necessary to forecast tourism demand accurately, but now there is no a centralized normal form for which kind of mathematics model can be used. At present mathematics models which based on statistics are often used to forecast, such as the time sequence model and regression model. This paper used the ANN which has been used extensively in many fields to forecast the inbound tourism demand of Zhejiang.Based on the theory of artificial neural network (ANN), this paper did a preliminary research on forecasting model selection, establishing procedure and achieving method of ANN of tourism demand. This model had used three-layer BP network to forecast the number of inbound tourists and foreign exchange earnings of Zhejiang. In order to make sure that the data is in the same quantity rank, the author adopts normalized method to treat the data imported and exported in advance when training BP net.