Research On A Enterprise Customer Order Prediction Based On IPSO Optimized LSTM Neural Network

With the gradual shift from mass customization to mass personalization in the manufacturing industry,manufacturing companies are experiencing highly volatile trends in their orders.Accurate order forecasting is of great significance for enterprises to improve operational management and economic efficiency.As a company facing intense competition and uncertainty,it is crucial for Company A to accurately predict customer orders in order to achieve an efficient supply chain and meet customer demands.However,traditional forecasting methods have difficulty producing satisfactory results due to the complexity and instability of order data.To address these issues,this study proposes an improved Particle Swarm Optimization-Long-Short Term Memory(IPSO-LSTM)method for order forecasting.Firstly,the IPSO algorithm is used to optimize the hyperparameter selection of the LSTM model to improve its predictive performance.Secondly,the LSTM model is trained using historical order data to learn the characteristics and patterns of time series.Lastly,based on the learned patterns and features,the model can predict future customer orders.To validate the effectiveness of the model,this thesis conducts empirical research using real order data from Company A.The experimental results demonstrate that compared to traditional forecasting methods,the IPSO-LSTM combined forecasting model significantly improves the accuracy and stability of order forecasting.The model achieves an order prediction accuracy of 86.8%,which is a 31.8%improvement compared to traditional forecasting approaches.Furthermore,through comparative analysis,the results indicate that the model exhibits better adaptability for order forecasting of low-voltage frequency converters for Company A.In conclusion,the IPSO-LSTM combined forecasting model can serve as an effective tool for customer order forecasting in Company A.The model not only provides accurate order predictions but also helps optimize supply chain management,enhance customer satisfaction,and achieve sustainable development for the enterprise.Future research can further explore the application potential of this model in other industries and domains,and carry out more in-depth optimization and improvement.