Choosing the appropriate order in fuzzy time series: A new N-factor fuzzy time series for prediction of the auto industry production

Abstract

In this paper, a new fuzzy time series based on high-order fuzzy logical relationships and Tabu Search is presented. The proposed method constructs N-factor high-order fuzzy logical relationships based on the historical data and uses Tabu Search and a parametric fuzzy inference system to adjust the length of intervals in the universe of discourse for prediction to increase the forecasting accuracy rate. We have applied our model for different cases with different factors. The model is applied for prediction of auto industry production of Iranian companies with a three-factor fuzzy time series model. The results show that the proposed method gets a higher forecasting accuracy rate than the existing methods in all cases.