Predicting tourism demand using fuzzy time series and hybrid grey theory.

Abstract Forecasting tourism demand in a capacity constrained service industry has been a major theme in this field. This study presents two models that can be used to predict tourism demand. Both two models are based on artificial intelligent (AI). Neural network theory was first applied to tourism demand forecasting in 2000 and empirically tested using the raw data from Hong Kong. This work provides empirical evidence using grey theory and fuzzy time series, which do not need large sample and long past time series. These AI models are estimated for tourist arrivals to Taiwan from Hong Kong, United States and Germany during the period of 1989–2000. GM(1,1) model achieves an accurate forecast when the sample data show a stable increase trend. Nevertheless, the Markov modification model can efficiently improved the GM(1,1) model when the sample data show significant fluctuations.