Design And Implementation Of Fertilizer Sales Analysis And Forecasting System

The analysis and prediction of fertilizer sales are helpful for fertilizer production enterprises to make production and operation decisions and better improve their own economic benefits.The traditional fertilizer sales forecast mainly relies on the experience of sales personnel in fertilizer production enterprises,and lacks scientific analysis of sales data in previous years,resulting in low efficiency,time-consuming and laborious problems.In recent years,with the continuous application of deep learning in agriculture,more and more deep learning prediction models have been proposed and applied.The time series prediction model based on deep learning makes full use of the fertilizer sales data of fertilizer production enterprises in previous years,which can achieve accurate prediction of fertilizer sales,provide support for the future sales forecast of fertilizer,provide decision-making basis for fertilizer production enterprises,reduce production costs and improve product profitability.This paper studies the prediction method of fertilizer sales data based on deep learning,and designs and implements a fertilizer sales analysis and prediction system based on deep learning for enterprises.The main research work of this paper is as follows :(1)The fertilizer sales data set was constructed.The fertilizer sales data of a famous fertilizer production enterprise in Shandong Province from January 2019 to December 2022 were collected,and the outliers and missing values of the original data were supplemented.Aiming at the prediction of fertilizer sales volume of enterprises,the impact factors were added,and the fertilizer sales data set was sorted out.The data totaled more than 40,000,which provided strong data support for the fertilizer sales prediction model.(2)Aiming at the problem that Transformer does not consider the context temporal correlation information of fertilizer sales data,resulting in low sales prediction accuracy,a fertilizer sales prediction model based on Transformer-BiGRU is proposed.Firstly,the multifeature extraction of fertilizer sales data is performed by Transformer,and the feature vector is output and input into the BiGRU layer to mine the context temporal correlation information of fertilizer sales data and realize the prediction of fertilizer sales.The results of comparative experiments show that the mean square error of the proposed algorithm is 0.08.On the same test set,the improved model has better sales prediction accuracy than Transformer and BiGRU.(3)A fertilizer sales analysis and prediction system was designed and implemented.The system consists of fertilizer sales analysis and prediction APP and fertilizer sales data management system.Fertilizer sales analysis and forecasting APP includes fertilizer sales data visualization analysis,fertilizer sales forecasting and fertilizer product introduction functions.Fertilizer sales data management system includes fertilizer sales data management and user information management functions.The system can assist enterprises in production and management decisions.