Online Order Priority Evaluation Based on Hybrid Harmony Search Algorithm of Optimized Support Vector Machines

Abstract

To make production plan, online order priority evaluation is the current priority weakness of online order evaluation model. This thesis proposes an online order priority evaluation model based on hybrid harmony search algorithm of optimized support vector machine (HHS-SVM). Firstly, an online order priority evaluation index system is build, and then support vector machine is adopted to build an online order priority evaluation model; secondly, harmony search algorithm is used to optimize the parameters of support vector machine; what's more, in the parameter optimization process the foraging behavior of AFSA is introduced to improve the ability and convergence speed of algorithm escaping from the local optimal solution; finally, simulation test is used to test the performance of the model. The simulation results show that HHS-SVM improves the accuracy of the online order priority evaluation relative to the comparison model. What's more, the online order priority evaluation model is feasible and effective.