基于聚类的钢铁生产成本状态分析方法

Production Cost Status Analysis Method of Iron and Steel Based on Clustering

面向订单生产的钢铁企业由于品种多、工艺复杂,不同产品、生产工序的实际成本构成要素是不同的,因此在实际产品成本的形成过程中存在复杂多样的产品生产成本构成状态,如何对其分析,并根据成本控制指标进行多维度成本控制是企业实现精细化成本事中管理的关键。提出基于控制和构成维度的生产成本状态描述方法,实现从多维度、多特征的角度对成本状态进行描述。建立基于K-means成本状态分析模型,通过对初始聚类中心、成本状态聚类距离的设计,提高了聚类结果的精度,实现了依据目标成本、标准成本、质量特征等对产品、作业等成本状态的量化分类。最后通过实例验证了该方法具有较高的分类精度,为面向实际生产作业在成本状态很难量化控制情况下,准确进行动态生产成本分析提供方法支持。

In make-to-order iron and steel enterprises with various products and complicated processes,real cost elements differ in products and production processes.So in the formation of factual product cost,there have been multiple forms of product cost status.How to analyze the complex production status and make multi-dimensional cost control according to cost control indexes is an urgent problem to be solved to realize enterprises' fine cost management in progress.The description method of product cost status based on cost status dimension'control and constitution is studied to describe the cost status from multi-dimensions and multi-features.The cost-status-analysis model based on K-means is presented to achieve quantitative classification management of cost status with target cost,standard cost and quality characteristics,etc.Through studying the initial clustering center and clustering distance of cost status,the model'accuracy is increased.Finally,the above method is verified to have better classified accuracy through the examples.It provides a methodological support for dynamic production cost analysis when there is difficulty in quantifying the control of cost status in real operations.