Reliability-Based Optimization Of Logistics Networks For Fresh Agricultural Products

Logistics system is a complex dynamic system, which composed of a number of functional aspects of mutual dependence and mutual influence. The normal operation of the system in various areas will be subject to many uncertainties, which leading to instability in the whole system performance. For fresh agricultural products, sometimes the unreliable of logistics systems not only lead to dissatisfaction of customers, but also led to huge losses. However, the current planning and design of logistics network take little customers'demand of system reliability. In addition, there are differences for the connotation and extension of the concept of system reliability in different application areas due to the difference of system functions and features. For the reliability of logistics system for fresh agricultural products, there is no clear definition yet, but also the lack of appropriate measurement indicators.Therefore, the meaning, failure mechanism and measure indexes of the reliability of logistics system for fresh agricultural products are proposed on the basis of the analysis of logistics models, characteristics under different models and network structures. And the relationship between logic structure and the reliability is analyzed. Furthermore, the reliability technology is applied into the planning of logistics system, and the reliability-constrained optimization models and relative algorithms for logistics network of fresh agricultural products are proposed to reduce the logistics depletion, logistics costs and operational risks, as well as to protect food safety and give full play to the economic benefits of fresh agricultural products..Concretely, the innovative works are as follow.1. The concept of logistics system reliability of fresh agricultural products has been defined from the system functions and features. And the failure modes in logistics system for fresh agricultural products have been presented:delivery delay, deterioration of goods, the number of delivery error, category error and location error. Among them, for fresh agricultural products, the delivery delay and deterioration of goods are the most common and serious faults.2. Considering customer's requirements for the timeliness of delivery and the degree of freshness when fresh agricultural products reach respectively, two evaluation indicators of system reliability, the "reliability of logistics time" and "reliability of quality" are defined and calculated.The existing literatures on the impact of time on the reliability of logistics systems only consider the transportation time between logistics nodes, without considering the time of other logistics operations (such as storage, packing, loading and unloading, material handling, distribution processing, etc.). In this paper, "reliability of logistics time" is defined as the probability of meeting customer's requirements for the total time, i.e. logistics time, spent in the logistics process of the fresh agricultural products from farm to fork. In addition, the decline of product quality in the logistics process of farm-products not only relates to logistic time, but also to many other logistic properties, such as ambient temperature, relative humidity, the coating and the gas content in packaging or containers. So it is difficult to comprehensively study the reliability of logistics systems only using the indicator of "reliability of logistics time". However there are no appropriate indicators and methods of measurement in existing literatures. Freshness is an important indicator to measure the quality of fresh agricultural products. In this paper, fresh degree is used as an evaluation standard to determine whether the logistics service for fresh agricultural products is failure or not, and the concept of "reliability of quality" is proposed. And considering the randomness of the failure of logistics services and the fuzziness of the description of ultimate consumers to fresh degree of agricultural products, the fuzzy-random reliability index of logistics service for meeting the requirements of freshness is defined and calculated.3. The optimization model and algorithm based on the logistics time reliability of logistics network for fresh agricultural products are established.Customer reliability of the logistics requirements of the time constraints as a confidence level of reliability and at the same time take into account the amount of loss occurred in the conservation of logistics constraints, the node capacity constraints, as well as the number of new facilities constraints of time based on the reliability of the logistics of fresh agricultural products Logistics network optimization of random chance-constrained model; structures based on stochastic simulation and genetic algorithm hybrid intelligent algorithm for solving the model, and validated through numerical simulation model and algorithm.The random chance-constrained programming model of logistics network optimization for fresh agricultural products is established, which take the customer requirements of logistic time reliability as the confidence level of reliability constrain, and at the same time take into account the logistic quantity constraints with logistics loss, node capacity constraints, as well as the number of new facilities constraints. And the hybrid intelligent algorithm for solving the model based on stochastic simulation and genetic algorithm is constructed. Numerical simulation results demonstrate that the model and algorithm is effective.4. The basic framework of the closed supply chain for fresh agricultural products is presented. On this framework, the optimization model and algorithm based on the quality reliability of logistics network for fresh agricultural products are established.The research results indicated that the closed supply chain is a kind of supply chain management model based on the quality and safety control, advanced information technology and modern logistics technology, which aim is to achieve the quality management of product. Therefore, in order to build a logistics network of closed supply chain for fresh agricultural products, it is necessary to take the quality reliability of logistics systems into account. The basic framework of the closed supply chain for fresh agricultural products is presented in this paper. On this framework, the fuzzy-random chance-constrained programming model of logistics network optimization for fresh agricultural products is established, which take the customer requirements of quality reliability as the confidence level of reliability constrain. And the hybrid intelligent algorithm for solving the model based on fuzzy stochastic simulation and genetic algorithm is constructed. Numerical simulation results demonstrate that the model and algorithm is effective.