Inventory demand forecast based on gray correlation analysis and time series neural network hybrid model
| Autor: | Fengjiao Cheng, Ruoying Sun |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Artificial neural network
Computer science 05 social sciences 020206 networking & telecommunications 02 engineering and technology Demand forecasting computer.software_genre Data modeling 0502 economics and business Correlation analysis 0202 electrical engineering electronic engineering information engineering Data mining Time series Hybrid model computer Gray (horse) 050203 business & management |
| Zdroj: | ICNC-FSKD |
| DOI: | 10.1109/fskd.2017.8393167 |
| Popis: | With the rapid change of market and the continuous growth of personalized need, the traditional forecasting methods are difficult to meet the requirements of enterprises. To solve this problem, this paper proposes Gray correlation analysis and time series neural network hybrid model. This model adopts the gray correlation analysis method that selects the main influencing factors as the input data of time series neural network. The time series neural network introduces delay module and output feedback module which not only considers the input and output in the past, but also has feedback ability. Compared with the traditional prediction method and BP network, the experimental results show that the accuracy of the model is 93.54%, significantly higher than general prediction method and BP neural network. So this model can be better applied to inventory demand forecasting. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|