论文部分内容阅读
The back-propagation neural network ( BPN ) is used to explore what the most influential variables are to drive business and leisure air passenger travel from Japan to Taiwan.The variables are systematically identified , evaluated and analyzed in detail.The results reveal that some factors affect both leisure and business air passenger transport , and the others only affect one of them.Flights from Tokyo to Taipei and average hotel rate in Taiwan are the two most important factors for forecasting the demand of leisure air passenger transport , while variables related to business activities have more effect on the demand forecast of business air passenger transport.By using BPN , a forecasting model that considers actual market segments is established , and the results show that it is an accurate tool to forecast air transport demand.
The back-propagation neural network (BPN) is used to explore what the most influential variables are to drive business and leisure air passenger travel from Japan to Taiwan. The variables are systematically identified, evaluated and analyzed in detail. The results reveal that some factors affect both leisure and business air passenger transport, and the others only affect one of them. Flights from Tokyo to Taipei and average hotel rate in Taiwan are the two most important factors for forecasting the demand of leisure air passenger transport, while variables related to business activities have more effect on the demand forecast of business air passenger transport.By using BPN, a forecasting model that considers actual market segments is established, and the results show that it is an accurate tool to forecast air transport demand.