Sustainable manufacturing considers that efficiently controlling the early stages of a manufacturing system is an important issue for industry enterprise. However, the data collected at this point is usually limited due to cost and time issues, making it difficult to realize the real situation in the production process. One of the ways to solve this problem is to use a small data build prediction system, such as the various grey methods. Although the grey forecasting model has been successfully adopted in various fields and demonstrated promising results, it can still be further improved. Meanwhile, Chinese bicycles have had a tremendous effect on society. This paper thus combines Chinese bicycle industry proposes a new way to improve forecasting accuracy. The new prediction model is called Rolling-TDE-GM(1, 1). The grey model contributes to forecasting the future demand, and the TDE method is applied to optimize the parameters of grey model based on the minimization of forecasting error. Also, the rolling mechanism can keep the real-time data. In this experimental study, the public dataset and a real case are used to confirm the effectiveness of the proposed model; the experiment indicates that the Rolling-TDE-GM(1, 1) can significantly improve prediction precision.