Research On Complexity Of Logistics Network Systems

Logistics network is not only an important manifestation of logistics activities, but also is an important indicator to measure the effectiveness of logistics activities. Logistics network system is a complex network of dynamic systems, which is a subset of the complex network system, so it has most of the characteristics of complex network systems. Complex network theory and method can be used in-depth analysis and accurate research objective laws on logistics network system operation, development trends and patterns of dynamic structure and function as well as the logistics network system. Meanwhile the cost of logistics network system is well forecasted and analyzed under the paragraph, daters, and models and so on.Logistics network system has the characteristics of small world network,:in logistics network, the average path can not only represent product delivery to the customer’s time but also the cost of the distribution products or intermediate products to the customers; Logistics network is the average clustering coefficient of logistics network node degree of mutual connection and exchange between enterprises; Logistics network, the core enterprise usually have one or more, many of the node enterprises around the core enterprise establish system of production, marketing, inventory, distribution network, greatly reflects the scale-free complex network.In the real logistics network, there are generally a large number of nodes, the topological structure and a huge number of nodes are shown "emergent" network evolution regularity and characteristics of the network dynamics. Logistics nodes in the network is influence each other, interrelated, and gradually expand to the mutual connection and mutual influence between different logistics network, interaction, with the evolution of complex mode of coupling between different network to promote, to form a complicated world. In addition, the dynamic logistics network system owes characteristics of self-organizing and chaotic.From elements of logistics network system components, associations between aspects of hierarchy elements, studies the characters of logistics network system while describes the functions of the logistics network system. This article analyzes the aims and methods of complexity of the logistics network planning. Finally, makes some research of the complexity on system prediction, while points out that there are several factors which are dynamic, non-linear, random, sudden etc.; this paper made its forecast of demand, cost and technology under the BP network. In short, raw data is used to get a reasonable BP network model by training operation; several data is inputted to predict for the final output.There are three new ideas as follow:firstly Complex network theory and methods is provided a new perspective for the study of logistics network system; secondary BP network model is tested by the data from enterprise M and the network errors are be calculated by the code in Matlab; the questions in logistics network system planning are analyzed by the views and ways of System Theory.