【作者】 邹毅峰；

【导师】 谢如鹤；

【作者基本信息】 中南大学 ， 物流工程， 2009， 博士

【摘要】 针对食品冷链物流系统研究中安全性量化指标缺乏的问题,本文借助食品科学和系统可靠性等相关理论和方法,构建了食品冷链物流的安全可靠度模型,并将模型应用于食品冷链物流系统的设计、评估和优化等方面。(1)食品冷链物流的安全可靠度模型。以微生物生长动力学模型和致病菌剂量-反应曲线为基础,借鉴系统可靠性理论,提出食品冷链物流安全可靠度概念,建立食品冷链物流单元的安全可靠度模型R=R0-d△T2t和食品冷链物流系统的安全可靠度模型Ri=R0-d(?)△Tj2tj。模型表明：冷链单元的失效率与食品经过冷链单元的时间成正比,与冷链单元的温度的平方成正比；冷链系统各单元的安全可靠度是相互独立的,且只与本单元的温度和时间有关,与系统初始状态和其他单元无关；在冷链系统中,安全可靠度从起始点到目的地都不可能增加,其安全可靠度下降是累积的,而且具有不可逆性。通过巴氏杀菌乳的冷链实验验证模型的可靠性。实验结果表明：食品冷链物流的安全可靠度模型在一定的条件下可以用来表征食品冷链物流的安全性。(2)食品冷链物流系统安全可靠度分配设计。以食品冷链物流系统安全可靠度模型为基础,利用温度与制冷系统能耗之间的函数关系,构建无成本约束和有成本约束的食品冷链物流系统优化决策模型,并给出了求解的启发式算法和算例。并以冷链物流单元的安全影响因子作为其重要度,进一步研究了食品冷链物流系统的安全可靠度再分配策略。(3)基于安全可靠度的食品冷链系统分析和评价。利用安全可靠度的概念和模型,确定冷链系统的关键单元、食品货架期和保存期以及利用GO-FLOW法进行冷链物流系统的安全评价。通过安全可靠度模型的灵敏度分析,得出了安全影响因子最大的单元就是影响冷链系统安全性的关键单元的结论,并利用排序法确定影响系统安全的关键单元。安全可靠度模型为预测食品的货架期和保存期提供了一种新的方法。巴氏杀菌乳保存期实验表明,温度的上升导致保存期快速缩短,销售环节是影响巴氏杀菌乳冷链物流安全的关键环节,要保证巴氏杀菌乳的销售安全,必须控制冷藏陈列柜的温度保持在10℃以下,并且外层的牛奶和里层的牛奶最好能每天调换位置。(4)基于安全可靠度的食品配送优化。比较分析两种主要食品配送模式(定量配送和定时配送)的安全性和经济性,提出集约化配送模式选择方法。研究表明对于服务水平要求高的食品,采用定时配送较好,而对于时效性要求不强的食品采用定量配送较好。将食品冷链物流安全可靠度引入车辆路径优化问题,以配送综合成本最低和安全可靠度损失最小作为目标函数,建立模型并利用最大最小蚁群算法进行了求解。以巴氏杀菌乳配送为例,利用安全可靠度,研究了食品配送时间、配送温度、装载堆码以及收货环节的优化问题。研究表明：巴氏杀菌乳的配送时间以不超过12h为宜；装载时尽量避免堆码在车辆的最里面,并且保持冷气顺畅循环；目前收货作业是影响配送安全的主要环节,应尽量缩短收货作业时间,以及消除冷链“断链”现象等。(5)基于安全可靠度的食品库存策略优化。主要针对食品的订货和出货模型进行优化,得到最优的库存控制策略。订货策略方面,利用安全可靠度模型构建生鲜食品的过期罚函数,通过对订货模型的比较分析,得到以下食品订货策略：一、安全性要求越高的食品订货周期要求越短,在同样的库存条件下,越容易腐败变质的食品其订货周期要求越短；二、需求率越大的食品订货周期要求越短,对于需求量较大的生鲜食品,必须采取高频率的订货策略,而不能采取大批量、小频率的订货策略。反之,对于需求量较小的食品,可以适当提高订货周期；三、库存条件越好的食品订货周期可以越长。出货策略方面,通过模拟仿真,比较了先进先出(FIFO)制度和最小安全可靠度先出(MSFO)制度的安全性。结果表明最小安全可靠度先出策略在安全性方面明显优于先进先出策略。更多还原

【Abstract】 Aiming at the problem of lacking safety quantitative index in the research of food cold chain logistics system, based on related theories and methods of food science and system reliability, this dissertation builds up the safety reliability models of food cold chain, and applies these models to the design assessment and optimization of food cold chain logistics system.(1) Safety reliability models of food cold chain. It puts forward the concept of safety reliability of food cold chain, and constructs the safety reliability models of cold chain logistics unit (R= R0-d△T2t) and cold chain logistics system (Ri= R0-d(?)△Tj2tj), based on microbial growth model and dose-response curve. These models show that the failure rate of cold chain unit is proportional to the time and the square of temperature; the safety reliability of cold chain units are mutually independent, it only relats to their respective time and temperature yet unrelated to initial state of system and the other units. The system model also reveals that the safety reliability of cold chain system reduces gradually and irreveribly. Finally, it verifies the models by pasteurized milk cold chain experiment; the results indicate that this model can represent the safety of food cold chain.(2) The allocation design of safety reliability of food cold chain system. Non-cost constrained optimization model and cost constrained optimization model have been built on the basis of safety reliability model and the function of temperature and energy consumption. Moreover, a new heuristic algorithm has been designed to find the solution. Further, it studies the reallocation dicision of safety reliability of system. Safety factors of cold chain units are in accordance with their component importance.(3) Safety analysis and assessment of cold chain system based on safety reliability. The concept and model of safety reliability can be applied to determine the key unit of cold chain, shelf life and use-by date of food. It also can assess the safety of cold chain logistics system based on GO-FLOW method. With the sensitivity analysis of safety factors of food cold chain system, it draws a conclusion that the unit having maximum safety factor is just the key unit of cold chain. The safety reliability model also provides a new method to predict the shelf life and use-by date of food. Based on pasteurized milk cold chain experiment, it is known that use-by date of food is quickly shortened with the temperature rising and the sale is the key safety unit. It is necessary to control the temperature of refrigerated display cabinet not exceed 10℃and outer milk and inner milk should change position every day.(4) Food distribution optimization based on safety reliability. Through a comparative analysis of fixed quantity distribution and fixed time distribution, some intensive distribution operation strategies have been provided. Fixed time distribution model is better than fixed quantity distribution model for those foods that require high service level, however, fixed quantity distribution model is better than fixed time distribution model for those foods with low timeliness. A model of vehicle route problem with time window based on safety reliability has been built, and solved using max-min ant colony optimization algorithms. Used the safety reliability model and dateof pasteurized milk experiment, it studies distribution time, distribution temperature, stacked and reception of food. The results show that the milk temperature at the doorway and in the middle of carriage is lower than the milk temperature in the innermost of carriage. So, milk should not be stacked in the innermost of carriage but loaded at the doorway and the middle of carriage; distribution time of milk should be limited to no more than 12 hours; reception time should be shorten as possible and cold chain breaking must be avoided.(5) Food inventory optimization based on safety reliability. By means of safety reliability of fresh food, a food ordering model integerated with penalty function of expiration is put forward to study the ordering policies. A. The higher food safety requirement is, the shorter food ordering cycle needed. Perishables require the shortest ordering cycle under same storing condition. B. The larger food demand rate is, the shorter ordering cycle needed. Fresh food at high demand rate must take high frequency ordering rule. Yet, it is proper to prolong ordering cycle for those at low demand rate. C. The better food storing condition is the longer ordering cycle could be. As regards out-stock strategy, the safety of two rules, which are first in first out (FIFO) rule and minimum safety reliability first out(MSFO) rule, have been compared by the means of simulation. The conclusion shows that minimum safety reliability first out rule is much safer than first in first out rule.更多还原