Demand forecasting method with enterprise productivity and demand state combined

Abstract

The invention discloses a demand forecasting method with enterprise productivity and a demand state combined. According to demand characteristics of a target group, a demand information model of the target group is built, and a demand questionnaire database is built; an enterprise productivity comprehensive adjustment coefficient is calculated; an evaluation set under the demand questionnaire database is used to calculate an interaction matrix, a demand state matrix is acquired according to demand standard threshold comparison, a state transfer matrix is calculated by using enterprise historical data, a forecasted demand state matrix is forecasted and acquired, and a demand state adjustment coefficient matrix is calculated; and an initial demand importance degree is calculated, a grey prediction method is used to calculate and directly forecast the demand importance degree in a next stage, and in combination of the enterprise productivity adjustment coefficient and the demand state adjustment coefficient, the final demand importance degree of the target group is calculated and acquired. The result for direct demand importance degree forecasting is further processed, and in combination of the demand importance degree value of the condition of the enterprise itself, the demand importance degree can be reflected more intuitively.

Description

technical field [0001] The invention relates to a demand forecasting method, in particular to a demand data forecasting method combining enterprise profitability and demand status. Background technique [0002] As a part of the development of future industrial intelligent manufacturing, demand forecasting has received extensive attention. The existing demand methods are mainly: investigate the needs of the target group, and distinguish the importance of different demand attributes, use the triangular fuzzy function to defuzzify the demand, and then use the product quality house to construct the demand-technology matrix to calculate the technical importance. Finally, use forecasting models such as autoregressive moving average model, gray forecasting, Markov chain, neural network, etc. to predict the importance of demand or technology, so as to provide directional guidance for the next stage of design. [0003] At present, the existing forecasts all focus on finding the law ...

AI Technical Summary

Problems solved by technology

[0003] At present, the existing forecasts are all focused on finding the law of demand data, expecting to predict demand results with higher accuracy and smaller errors, but paying too much attention to the data itself but ignoring the impact of the status and type of each demand on the forecast results

Images