支持向量回归在不确定面向订单装配环境下交货期预测中的应用

针对面向订单装配(ATO)生产环境，如何根据订单信息、生产系统特性快速地估算出准确、可靠的交货期问题，在分析不确定性要素对交货期影响机制的基础上，构建了订单交货期预测模型。模型参数包括三个部分：订单上线时间、装配周期和异常拖期。订单上线时间基于零部件、生产能力的可用性，订单装配周期和异常拖期采用基于实际生产历史数据的支持向量回归(SVR)方法进行预测。案例研究表明该模型预测结果与实际交货期接近，可以用于指导订单交货期协商。

Application of support vector regression in prediction of due date under uncertain assemble-to-order environment

For the issue of how to quickly estimate the accurate, reliable due date according to the order information and the features of the production system in Assembly To Order (ATO), a due date prediction model was constructed based on the influential mechanism analysis of the uncertainty factors. The model parameters included three parts: order release time, assembly cycle time and abnormal tardiness. Order release time was based on the availability of materials and production capacity. The assembly cycle time and abnormal tardiness were predicted by using Support Vector Regression (SVR) method based on actual production history data. The case study shows that the predicted results of the model are close to actual due date and it can be used to guide the order's delivery time consultation.