汽车制造业物料供应需求优化预测仿真研究

Auto Manufacturing Material Supply Demand Forecast Simulation Optimization

对汽车制造业物料供应需求进行准确预测,可对汽车制造业物料供应链进行优化,降低供应链的整体运作成本。进行物料供应需求预测时,应计算当前生产对汽车制造业物料的需求量,但是传统方法只能凭经验初步获取汽车制造对物料的需求量,根据需求量建立物料供应需求预测,不能准确计算当前生产对汽车制造业物料的需求量,导致建立的模型存在较大误差,不能对物料供应需求进行准确预测。提出一种遗传算法与改进BP网络的汽车制造业物料供应需求优化预测方法。上述方法先建立了泊松分布的物料需求时间模型,确定当前生产对汽车制造业物料的需求量,对汽车制造业物料需求时间序列进行相空间重构,利用BP神经网络构建物料预测样本矩阵,融合遗传算法优化BP神经网络的权值和阈值,获取汽车制造业物料需求预测的最优解函数,以上述函数为依据对汽车制造业物料需求进行预测。仿真结果表明,所提方法为汽车制造企业入场物流的有效组织提供了依据。

An optimization prediction method of material supply demand in the automotive manufacturing industry is proposed based on a genetic algorithm and an improved BP network. A material demand time model with Poisson distribution is established,and then the demand for materials in the automotive manufacturing industry is determined.The phase space of material demand time series is reconstructed. The BP neural network is used to build the sample matrix to integrate the genetic algorithm to optimize the weights and thresholds of BP neural network. The optimal solution function of the material demand prediction is obtained and used to achieve the material demand prediction of automobile manufacturing industry. The simulation results show that the proposed method provides a basis for the effective organization of entrance logistics for automobile manufacturing enterprises.