Demand Forecasting Model For Short Life Cycle Products Based On Improved Bass

The continuous change of the consumer’s preference leads to the improvement of the product’s innovation speed. The shortening of the product life cycle has become a trend. The enterprises that produce short life cycle products are paying more and more attention to the demand forecast, which is expected to guide the actual production of the enterprises with more accurate prediction. As the seller’s market is gradually changing to the buyer’s market, the demand forecasting of short life cycle products has important theoretical value and practical significance.The core problem of this paper is: how to build a short life cycle product demand forecasting model scientifically and systematically? The questions include:(1) how to make use of the similar product data to make up for the lack of data in the initial stage of product development?(2) How to build a model of the short life cycle product demand forecasting?(3) How to improve the short life cycle product demand forecast accuracy?Based on the analysis of relevant research at home and abroad, this paper establishes an improved BASS model which is suitable for short life cycle product demand forecasting. From two aspects of how to solve the lack of initial data and how to improve the accuracy of prediction, the key issues such as the measurement of product similarity, the consideration of the influence factor of the demand and the solution of the parameter rolling update are studied deeply. The specific research contents include:(1) describes the fuzzy c-means clustering algorithm and rough set theory, system similarity principle based on, is proposed based on the characteristics and the importance of the objective measure product similarity method, solve the problem of short life cycle product data in the early stage of deficiency leads to the prediction accuracy is not high.(2) In view of the BASS model does not consider the diffusion process of random factors, establishes an improved BASS model adding consumer preference factors and price factors.(3) Aiming at the lack of adaptability to the model leads to the problem of low accuracy of prediction, a new parameter rolling update estimation based on Bayesian updating is proposed in order to improve the short life cycle product demand forecast accuracy. For short life cycle products, this paper hopes to put forward a set of system of product demand forecasting method, in order to guide the enterprise to the product demand forecast.In this paper, we take the mobile phone demand data of a company as an example to verify the validity of the model and the algorithm.The research methods of this paper mainly include survey method, statistical processing method and fuzzy comprehensive evaluation method. Among them, the weight of similar products is designed by using fuzzy clustering rough set method, the model parameters are solved by statistical processing method, and the data are obtained by questionnaire survey method and literature research method.