A method for forecasting the demand of emergency materials considering partial data ambiguity and missing

The invention discloses a method for constructing an emergency material demand forecasting model considering partial data fuzziness and missing and forecasting the emergency material demand based on the model, Aiming at the problem that the turning point of whitening weight function is difficult to determine the fuzzy data, Two formulas for calculating the 'kernel' of fuzzy interval grey numbers are presented, The obtained 'kernel' replaces the original fuzzy information to achieve the purpose of converting the uncertain information into definite information. Aiming at some missing data, an improved GKNN algorithm is proposed based on the combination of gray relational degree and K-nearest neighbor filling algorithm, in which weights are introduced in the filling part and logic checking conditions are added after filling. Then input the preprocessed data into the improved genetic algorithm optimized neural network model to get the trained emergency material demand forecasting model, and test the forecasting m