Combined grey model-based prediction management method for raw material requirements of curl tobacco leaves

Abstract

The invention relates specifically to a method for predicting and managing demand for recured tobacco leaf raw materials based on a combined gray model, and belongs to the technical field of tobacco inventory management. The method includes four steps: the establishment of the basic model, the calculation of preliminary forecast and error, model combination and weight calculation, and the forecast of output and necessary stock of raw materials. The method of the invention is simple and clear, easy to operate, does not need to increase equipment, does not change the existing production and management mode, can effectively guide and adjust the proportion of raw material procurement, and ensure the dynamic balance of supply and demand of raw materials in stock. The invention adopts a combination model to reduce the systematic error caused by a single model and improve the prediction accuracy; and not only can predict a single brand, but also can be used for total quantity prediction, so as to provide scientific and effective production and management decisions for producers.

Description

technical field [0001] The invention belongs to the technical field of tobacco inventory management, and in particular relates to a method for predicting and managing demand for recured tobacco leaf raw materials based on a combined gray model. Background technique [0002] The inventory management of re-cured tobacco leaves is different from the inventory management of other semi-finished products. First, it needs to be aged for a period of time before it can be used in production. Tobacco leaves are processed into red-cured tobacco through picking, drying, re-baking and other processes. Re-cured tobacco leaves cannot be directly used for cigarette production, but must be naturally aged. The style and quality of the aged re-cured tobacco will be better, and the taste of the produced cigarettes will be better. Secondly, the re-cured tobacco leaves from different origins have inconsistent requirements for the necessary storage time. Some scholars have done experiments befo...

AI Technical Summary

Problems solved by technology

However, there are some defects in the stochastic analysis of time series: First, the premise of the analysis is that the data can be transformed into a weakly stationary sequence after a finite number of difference transformations.

The tobacco industry is greatly affected by policies, and extreme values ​​caused by policy factors will make the prediction model not robust, with large errors in prediction values ​​and poor prediction accuracy

In addition, the trend extraction method of time series directly analyzes time as a variable. The defect of this method of forecasting by extracting the trend of things changing over time is that it does not consider the influence of external factors and ignore external impacts. It is considered to be stable over time, so there may be a large error in the prediction of this method

Regression analysis focuses on highlighting the quantitative causal relationship between variables, and there are some shortcomings when used for prediction: on the one hand, the data requirements are strict, and the prediction model assumes a lot

Therefore, it is difficult for the regression model to comprehensively identify the guiding factors that affect the raw material demand of re-cured tobacco leaves

In addition, the basic assumptions of regression analysis are too harsh, so the expected accuracy of this method is difficult to achieve the expected effect

Images