摘要
针对制造业产业链协同服务平台的备件业务协作需求,提出跨节点的库存协同解决方案并建立近期需求预测计算模型。结合分布式节点企业的历史交易数据、库存数据的实时采集与处理应用,保障库存控制方案的实效性。采用MapReduce框架对模型参数计算过程进行优化,提高运算速度。基于遗传算法获取模型计算最优解,并将模型计算结果推送至下游经销商企业群,由反馈信息控制订单的动态生成。并将该模式应用在汽车产业链云服务平台,压缩了产业链响应时间。
To meet the spare parts collaborative needs of manufacturing industry chain collaborative platform,we propose a cross-enterprise inventory control solution and establish a near-term demand forecasting model. In order to ensure the effectiveness of the inventory control program,w e dynamically extract historical transaction data of each node and real-time inventory data,and obtion based on the genetic algorithm. We use the MapReduce framework to calculate model parametersand improve the processing speed. By pushing the results to downstream service providers,the feedback information can control the dynamically generated orders. The application in the automotive industry chain cloud services platform demonstrates that the response time of the industry chain is compressed.
出处
《计算机工程与科学》
CSCD
北大核心
2017年第10期1812-1818,共7页
Computer Engineering & Science
基金
中央财政服务业发展专项资金(2015059901010)
四川省科技支撑计划(2014GZ0142)
关键词
备件
产业链
库存协同
遗传算法
spare parts
industrial chain
inventory cooperative
genetic algorithms
