Sales Forecasting Methords Of Short Lifecycle Products And Application In Zhanwang Electronic Company

With the rapid development of economic, the lifecycle products are getting shorter,especially for the high technological products such as electronic products.The aim to predict thesales of short lifecycle products is to save costs.How to improve the production efficiency is verysignificant. The short lifecycle products are also known as aging goods, perishable or seasonalgoods.The class of goods that is sold relatively in a short time. There are two meanings for shortlifecycle: On the one hand, the short lifecycle products mean the physical time not long.Thevalue of products will be declining sharply; On the other hand, it means the market lifecycle isshort. The market lifecycle is elapsed from the time that the product is saled from on the marketto the time to be out of the market. Predicting the sales of electronic products accurately means alot to the mamager. And it can become more simple and effective for the mamager to makedecision. Besides it also can reduce the inventory cost of the productions.BASS algorithm is commonly used for the prediction of short life cycle products and forresearch. This paperanalysed and comparised the existing prediction algorithms.BASS diffusionmodel in the prediction of a short life cycle products did not considered the eliminate noise.Andit ignored the issue of the impact of seasonal factors as well. This paper is proposed on the basedof wavelet analysis in order to remove noise.At the same, this paper seasonal factoris used toimprove the BASS model algorithm in order to predict the short life cycle of productsales.Then,the experiment was put on the Outlook e-company sales data.The result verified theimproved BASS model’s feasibility. This paper includes the following aspects:Firstly, after introducing a lot of literature, this paper introduced the concepts of a short lifecycle products both at home and abroad sales.The characteristics current algorithms Profile.Then,the advantages and disadvantages of the various algorithms is also mentioned in this paper.Secondly, a detailed study was carryed out to introduce the basic theory of the BASSdiffusion model, including the methods for the short lifecycle products, models of assumptionsand model building.then the BASS diffusion model’s limitations is also mentioned in this paper.As there are noise processing and no adjustment for product seasonal fluctuations impactprediction in BASS diffusion model, in Chapter IV, a careful study was carry out to be able tohandle this problem.Wavelet analysis theory and seasonal factor are used in this paper.In thispaper, it outlines the development process of the wavelet analysis method, inlcuding the basicconcepts and common methods. Seasonal factor is extracted including the basic theoreticalconcepts. BASS model is etracted by wavelet filtering and seasonal factors. Last, a design simulation was carried out in the Outlook electronic company.The actualsales data is used to show wheather the improved BASS diffusion model algorithm is reallybatter than the existing prediction.