Research On Weapon System Portfolio Selection Problems And Decision Methods

Joint operations and system of systems counterwork have become important development trends of modern warfare,Joint Together and Architecting are not only the goals of the weapon equipment development,but also the important concept influencing the planning pattern of the weapons and equipment development.During the weapon equipment development planning process,Decision-makers inevitably are confronted with a variety of choosing issue.Because the weapon system selection is closely related to weapons production,deployment and operations,which determine the result of the war,it attracts more attention from the military decision-makers and researchers.Traditional chimney pattern based on sector-centric development model and fragmented decision-making process greatly hamper the process of joint capability generation from individual weapon systems in the system of systems level,however,the pursuit of integration,universalization,serialization become distinctive feature of the current weapon equipment development planning.Therefore,the development of weapon equipment must emphasis more on weapon systems portfolio decision-making and management,considering various weapon systems as a whole for trade-off analysis and acquisition decision-making.Our study carries weapon system planning under the joint system of systems background,and unifies the weapon system selection decision-making process by portfolio decision-making theory creatively,then designing the problem framework and certain decision-making methods.The selection problem focuses on the production and deployment phase and operation and support phase,which is comprised of :1)weapon system portfolio selection for production;2)large scale weapon system portfolio selection optimization;3)weapon system portfolio selection to be production;4)weapon systems portfolio selection with critical condition.The main contents and innovations are given as follows:(1)A weapon systems portfolio selection model is proposed regarding co-effects from requirements pull and technology push.Military requirements pull and key technology push are two important driving forces for the weapon systems development into production.Therefore,from the above two dimensions,in the basis of Value Focused Thinking(VFT)concept,multiple criteria analysis is employed to build a portfolio value hierarchal structure from two main perspectives: a five-level set of technology pushing measures and a two-level set of requirement pulling measures.With regard to requirements satisfaction level,it is obtained by requirement pulling measures based on linear programming optimization and the first-ignorance model,which is founded on expert opinion,using pair wise comparisons.Then,the dominance structure of the portfolios is designed and all the technology and requirement portfolios(T&R portfolio)are generated as a set of optimal solutions satisfying certain constraints,through multi-objective integer programming model,maximizing the requirement satisfaction level as well as technology readiness level.Finally,cost sorting optimization strategy is employed to search the Pareto-optimal solutions for weapon system selection.(2)A multi-scenario-based weapon system portfolio selection model is proposed with system dependency.Under real or simulated operations background,weapon system portfolio selection aims to selecting a combination of weapons systems which has the best operational effectiveness/value,the highest robustness/the lowest risk under different scenarios and the lowest cost/consumption.Scenario concepts is proposed to simulate the future operational requirement,which is defined as a triple-tuple consisting of state,expression and state transfer.According to different information completeness,it can be divided into two types,Markov scenario under incomplete information and Hidden Markov scenario under uncertainty information.Especially for the uncertainty condition,the interval number theory is introduced to achieve the unknown scene state recognition.besides,the possible relationships among different weapon systems are analyzed and described as four models,which are following synergy,synchronizing synergy,exclusive synergy and independence synergy.Finally,a scenario-based multi-objective programming approach is proposed to solve the selection problem with relative optimization algorithm.(3)A weapon system portfolio selection approach based on intelligent evolutionary algorithm is proposed for large-scale complex system of systems.With regard to large-scale complex system of systems,the increasing candidate weapon systems for selecting can lead to a classical problem-NP hard problem.The time consumption of weapon systems portfolio selection using by sorting strategy algorithm is intolerable,so a portfolio selection optimization is necessary to carry out.First,the above weapon system portfolio selection models can be abstracted as two categories: single-objective model and multi-objective model,whose features are analyzed and feasible solution scale reducing strategy is designed regarding to the difficulties faced in optimization.Next,the differential evolution algorithm(DE)is employed as the main method for problem solving,combined with TOPSIS method for acquiring weapons systems empowerment in a certain portfolio,which is called as DE-TOPSIS selection and optimization algorithm in our study.Finally,for three different scale weapon systems situations,single-objective experiments are carried out by DE algorithms,compared with other two algorithms,PSO and GA,while multi-objective experiments are carried out by NSGAII and NSDE algorithm,respectively.In result,our DE-TOPSIS algorithm has best performance on all indices,such as convergence speed,stability or the quality of optimal solution.(4)A weapon system portfolio selection model is proposed with a preference baseline value from decision-makers.Multi-criteria model are employed for the above three weapon system selection model,and the three have a similar phenomenon,there exists some weapon system indifferent for decision-making when they decided to chose or discard it.Moreover,it is very important for decision-makers to set a baseline value when using the multi-criteria model,which determine the boundary between chosen weapon system and discarded weapon system.Based on the definition of redundancy system,baseline system and baseline value,the weapon system portfolio selection approach was proposed with regard to baseline system.Starting from the two core parts of the approach,selection stratergy analysis and weapon systems ranking,the stratergies analysis of weapon system selection under single-objective and multi-objective were carried out first.Then,the interval number theory was employed to extend the VIKOR method to E-VIKOR method,with linear programming model as the weighting method under the uncertaties,aiming to rank the candidate weapon system.Finally,the three different rankings of candidate weapon systems under three different weighting scheme.In the basis of weapon system ranking,baseline value and selection stratergy,the refining weapon system selection was developed.