A Study On Demand Forecasting For Products With Short Life Cycle

Nowadays, the technology is developing rapidly, the companies attach great importance to R&D and the customers'preferences are changing all the time, and thus the products'life circle are becoming shorter and shorter. Short life cycle products are a kind of special but important products. In order to meet the market demand and make production plan, the companies are paying more attention on short life products'demand forecasting. For demand forecasting, it is necessary to choose forecasting method, improve the original method or develop new methods according to products'demand characteristics. Traditional forecasting methods may be inappropriate for forecasting of short life cycle products, because they usually do not take into account the characteristics. Therefore, the problem about demand forecasting for products with short life cycle has significant value in theory and practice to enrich forecasting theory and company practice.This thesis summarizes the short life cycle products'demand characteristics. Based on these characteristics, the conclusion that short life products'forecasting is inevitable is got. Then, the concept of diffusion and the development of diffusion model are introduced. As the most widely used model, BASS model is discussed. The researches BASS model's structure, the model's application method and the validity of market diffusion are presented. After the application feasibility analysis, BASS model is applied to forecast the demand of short life cycle products. According to the demand characteristics, we complement the main information with the data of past products'. The model is revised with seasonal factor. The numerical illustration proves that the revised BASS model could generate good forecast for products with short life cycle. The forecasting error of MAD, RMSE and MAPE are better than other methods. Base on this study, the further research is put forward. They are dynamic model development study, multi-generation products forecasting study and demonstration study.