摘要
建立了一种含可调参数的白酒勾兑混合整数线性规划模型,通过调节可调参数,得到不同整坛白酒数目参与勾兑的方案。为模型的求解提出了一种分层求解策略,即内层采用单纯型法优化连续变量,外层采用列队竞争算法优化整数变量。将提出的方法用于白酒勾兑的模拟计算,并与线性规划法相比较。结果表明:由该方法得到的优化方案,其参加勾兑的总的白酒坛数更少、整坛的白酒坛数更多,能有效地提高白酒品质及原度酒的储存空间利用率,降低勾兑操作成本及库存费用。
A mixed - integer linear programming model with an adjustable parameter for liquor blending was established. The blending schemes with different number of full jars were obtained by the adjustable parameter. A hierarchical strategy was proposed to solve the model, i.e. using the simplex method for optimizing continuous variables inside the layer and using line - up competition algorithm for optimizing integer variables outside the layer. The method was tested for liquor blending problem. The results show that the optimal scheme obtained by the method have more number of full jars and less total number of jars than that by linear programming. The optimal scheme can reduce operating costs, improve liquor quality and effectively increase the space utilization for original liquor storage, reducing inventory costs.
出处
《武汉理工大学学报（信息与管理工程版）》
CAS
2008年第3期398-401,共4页
Journal of Wuhan University of Technology：Information & Management Engineering
关键词
白酒勾兑
混合整数线性规划
列队竞争算法
分层求解
liquor blending
mixed - integer linear programming
Line -up competition algorithm
hierarchical solution
