A novel nonlinear forecasting model for output of bike industry by Grey model and Taguchi-differential evolution algorithm

Global marketing status shows that over 120 million bikes are being produced around the world yearly. Taiwan applied a cross-strait division of labor to manufacture about 30 million bikes in 2009. Since many study applied forecasts are basic inputs for many kinds of decisions and policy in business and government. This paper proposed a novel nonlinear forecasting model, known as Taguchi-differential evolution algorithm grey forecasting model (1,1) (TDEGM(1,1)). The TDEGM(1,1) is validated using integrated circuit and bike industry forecasting as study cases. Furthermore, the mean absolute percentage error is used to compare the performance of the forecasted models. The TDEGM(1,1) was the most accurate one among these models which are original GM(1,1), the Bayesian GM(1,1) and genetic algorithms GM(1,1). Meanwhile, the convergence result of TDEGM(1,1) is proved faster than genetic algorithms-based Gm(1,1) and Bayesian GM(1,1). In the 2013, the output value of bike maybe over 2000 million according to forecasting result.