Synthesis Modeling Of Complex System Lifycycle Cost With Reliability Constraints Under Weibull Distribution

Complex system is confronted with dire challenges of increased design cost, production cost, maintenance cost and inventory cost. If production cost were still used as a basis for decision making, it will produce great deviation. Under such background, the US military put forward the concept of lifecycle cost in 1950. It can not only guarantee the performance of the system, but also get the most efficient use of limited resources and the optimal economical benefits.According to the Pareto curve and the characteristics of complex system, it is necessary to make lifecycle cost decisions from aspects of performance,economy, maintenance and guarantee in development phase. An integrated research into cost optimization, maintenance strategy, inventory security and system performance is required to make up the disadvantages of individual processing. A lifecycle cost optimization model is given by synthesizing all kinds of factors. In theory the trade-off between complex system performance and cost is achieved to ensure global optimization by the model.First, under the condition that the life of homogeneous components follows Weibull distribution, a group preventive maintenance strategy is introduced by maintenance decisions, and then a lifecycle cost optimization model under reliability constraints is given by considering its structure and cost characteristics. The global optimal interval can be obtained by the discrete curve consisting of locally optimal solution. The curve of optimal lifecycle cost with group maintenance strategy is compared with no group preventive strategy and the intrinsic relationship of reliabilitybased design, maintenance strategy and optimal lifecycle cost are revealed by a utility curve altogether. The application value of the proposed model and optimization algorithm are illustrated using an example.Second, for the generic two-echelon logistics support network, an operational availability model under Weibull distribution is built on the basic of the improved algorithm of the exponential method. And then a lifecycle cost optimization model under reliability and operational availability constraints is proposed to study the relationship of system performance, cost and stock level. An numerical example is provided to demonstrate the effectiveness of the proposed model. Some sensitivity analyses for key parameters in the model are also investigated to improve the decisionmaking accuracy.The model and its algorithm mentioned above help system engineers estimate lifecycle costs of alternatives form the development phase. It provides not only a reference for performance design of complex systems, but also a scientific basis for management decisions.