Applied Research on Logistics Demand Prediction Based on Support Vector Machine of Genetic Algorithm

As an advanced organization and management technique, modern logistics´ application has been the focus of the enterprise management. However, due to Bullwhip Effect, logistics demand information is often distorted, reducing efficiency of many sectors such as users, retailers, wholesalers, and manufacturers. To improve the efficiency of logistics activities and ensure the balance between supply and demand of logistics services, on the basis of comparison and analysis, this paper selects the appropriate predictor system and uses the genetic optimization algorithm for least squares support vector machine combined to create a logistics demand forecasting model. Evidence shows that this prediction method has higher prediction accuracy to have broad application prospects in the logistics demand prediction.