节点文献

炼油企业供应链中的生产—库存优化系统研究

Study on Production and Inventory Optimization System of Refinery Supply Chain

【作者】 陈宏

【导师】 何小荣;

【作者基本信息】 清华大学 , 化学工程与技术, 2005, 博士

【摘要】 供应链管理是通过协调某一行业上下游相关企业的运营,增强整个供应链的竞争力,强调的是整体利益与长期效益,注重的是双赢、多赢。因此将供应链管理理念引入炼油企业的生产管理过程中,有利于提高石化行业的管理水平,更好的适应市场需求,提升企业效益和竞争力。生产过程管理是炼油企业效益的主要来源之一,故本文的研究重点为炼油企业供应链中生产-库存管理优化模型的建立及求解算法,在以下几个方面取得了成果:(1)将供应链管理理念引入炼油企业生产计划优化中,为炼油企业供应链管理的研究提供了有益的尝试。(2)一般的炼油企业生产计划优化并没有同时考虑产品库存的优化,导致了企业存在巨额库存,由此占用着企业大量的流动资金,造成了效益流失,不仅影响到企业的运行效率,而且削弱了企业的核心竞争力。因此,本文提出了在炼油企业供应链管理中需要同时考虑生产计划和库存管理优化的思想,以取得更好的经济效益。(3)通过分析、对比离散工业和炼油企业的供应链库存管理特点,构建了炼油企业供应链的单周期生产-库存管理优化模型,并通过实例计算验证了模型的可行性和有效性。(4)通过分析、研究发现,在常用的生产计划优化模型中,由于采用生产装置的侧线收率作为决策变量,致使约束条件不够充分,这可能会导致优化结果中的某些侧线采出不满足产品质量要求。为了解决这个问题,本文提出在生产计划优化模型中以总馏出率(某条侧线至顶层塔盘的总馏出率)代替侧线收率作为决策变量。(5)建立了基于产品需求预测的炼油企业供应链多周期生产-库存管理优化模型,并且通过实例计算证明了该模型能够使企业同时达到平滑生产、增强抗风险能力和降低油品总库存费用三个目的。(6)根据炼油企业供应链的多周期生产-库存管理优化模型特点,采用遗传算法与线性规划相结合的方法,将求解非线性规划模型简化成求解线性规划模型,使模型的变量数目及约束方程数量分别减少了 41%和 54%,极大地降低了

【Abstract】 Supply chain management intends to enhance the competitiveness of thesupply chain through the cooperation and coordination of the interrelatedenterprises within the chain. It pays attention to the total benefit andlong-term benefit of the whole supply chain. Therefore, the introduction of theconcept of the supply chain management to the production managementprocess of petroleum refining enterprises is beneficial to enhance theirmanagement level and improve their economic efficiency and competitiveness.Management and optimization of production process is one of the mainsources of the economic benefit of the refineries. In this dissertation, theproduction and inventory optimization system of the refinery supply chainwas studied and the following progresses have been achieved.1. The concept of supply chain management was introduced into theproduction planning optimization of refineries. This provides a usefulexploration for the research of supply chain management of refineries.2. Generally, the inventory management was not included in theproduction planning optimization of the refinery. Thus, the enterprise alwaysholds large quantities of finished product storage, and the efficiency andcompetitiveness is seriously decreased. Therefore, remarkable economicbenefit can be obtained through simultaneously optimizing the productionplan and inventory management of refineries.3. By analyzing and comparing the characteristics of discrete industryand refinery industry, a single-period production and inventory managementmodel for refinery supply chain was proposed. And a case study was carriedout. The results show that the model is feasible and effective.4. It has been discovered that in the general production planningoptimization model, as the yield of side draws are considered as importantdecision variables, so their constraint equations are insufficient. Thus, someof the optimized side draws may not satisfy the quality requirements. Herenew variables, the total cutting out ratios (total cutting out ratio from theside-draw plate to the top plate), were proposed to substitute the originaldecision variables (yield of side draws) to improve the production planningoptimization model.5. A multi-period production and inventory management optimizationmodel, which is based on the prediction of product demands, was establishedfor the refinery supply chain management. A case study with industry datafrom a refinery shows that the smoothing producing, fulfilling customerrequest and decreasing the entire inventory cost was achieved.6. A mixed algorithm based on the genetic algorithm and linearprogramming was proposed to solve the multi-period production andinventory management optimization problem for refinery supply chainmanagement. Using the mixed algorithm, the multi-period model wassimplified from non-linear programming to linear programming, and thenumber of variables and constraint equations was decreased by 41% and 54%,respectively.

  • 【网络出版投稿人】 清华大学
  • 【网络出版年期】2006年 08期
节点文献中: