Multi-capability Areas Weapon System Portfolio Selection And Decision Making Methods

The demonstration and planning of multiple armaments strive to make an overall design of armaments construction and do demonstrations on the evaluation,selection,and planning under national development strategy and military mission.Throughout the decision and planning process,decision makers have to face a series of selection and planning problems.However,the essence and pattern of modern wars have changed significantly,adding more difficulties to military decision makers.On the one hand,the development of weapons generally involves multiple capability areas,thus,tradeoffs in terms of resources among multiple capability areas are necessary.Traditional resource allocation models ignore adversary's possible behaviors and base decisions only on specific values or one's own development,which may cause serious problems;On the other hand,traditional selection and decision making methods for a singular weapon cannot be applied nowadays.More attention should be paid on weapon portfolios and historical performances in certain periods need be considered to guide the selection process.In addition,multiple stakeholders engage in the demonstration and planning process,thus,how to negotiate various opinions,deal with possible divergence,and select an ultimate solution that satisfies all is of great significance.Traditional selection and planning methods have put a hard squeeze on accounts for weapon development,which greatly influences the fighting capability.Aiming at solving the difficulties of choosing and planning in the weapon portfolios analysis,a decision making and analysis framework for weapon portfolio solutions is proposed from a top-level perspective.In summary,this thesis carries out research on weapon resources allocation within multiple capability areas,weapon portfolios selection in each capability area,and group decision making and analysis,in order to select an appropriate weapon portfolio in each capability area.The main works and contributions can be summarized as follows:(1)A decision making and analysis framework for weapon portfolios selection and decision making regarding multiple capability areas is proposed.Aiming at solving various technology problems,there exists a lack of efficient analysis procedures and approaches in the weapon decision making process to date.In order to change traditional service-driven analysis and individual programs in terms of weapons selection and planning,this paper focuses on weapon portfolio solutions and proposes a decision making framework from a high-level perspective.The framework consists of three main works: 1)weapon resource allocation regarding multiple capability areas;2)weapon portfolio evaluation and selection;3)group decision making among weapon portfolios.Each work corresponds to a certain development level throughout the entire selection and planning process.Research is carried out on the selection of weapon portfolios,doing tradeoffs among weapon portfolios,and planning for the portfolio solutions.(2)A prospect theory based weapon resource allocation model regarding multiple capability areas is proposed.Traditional resource allocation models often neglected adversary's possible attack behaviors.This paper proposes a prospect theory based weapon resource allocation model regarding multiple capability areas.Different from traditional expected utility theory or risk probability analysis,prospect theory can successfully describe individuals' risk preferences in decision making.Research on how to transfer the outcomes of different events into prospects based on cumulative prospect theory is first studied.Then,probabilities of successful(failed)attacks are transformed into transfer decision weights.Next,a certain logit model is used to estimate adversaries' target choice probabilities.Last,a resource allocation model aiming at minimizing the expected overall loss is proposed,which can be used to guide the allocation and selection process within multiple capability areas.In addition,comparative analysis is carried out among other strategies,analyzing the necessity of taking into account adversaries' transfer behaviors in the allocation process.(3)A mean-variance theory based weapon portfolio selection model is proposed.Traditional weapon selection models based their selections only on weapon characteristics to date.This paper aims to select weapon portfolios based on weapons' historical performances in a certain capability area.More specifically,weapon characteristics in different periods in terms of various criteria are first collected,based on which a multi-attribute decision making method is used to evaluate these weapons.Motivated by the mean-variance model,three aspects of the weapons,listed as average value,development trend,and relations among weapons,are modeled as three main objectives to form a multi-objective programing model.To tackle the uncertainties in many real-world situations,interval numbers are used to describe weapon attributes.By considering multiple constraints,a mean-variance theory based weapon portfolio selection model is proposed,and more robust weapon portfolio solutions could be obtained by solving the model.(4)A group preferences based model for weapon portfolio decision making and analysis is proposedIn general,more than one stake-holder engage in the weapon selection and planning process.This paper studies a group preferences based model for weapon portfolio decision making and analysis.Making full use of different knowledge and considering various preference demands,the obtained decision results could be more realistic.To tackle the complexities in real world situations,incomplete preferences are allowed.The multiplicative transitivity property of preference relations are used to estimate the missing values.Then,a consensus reaching model is put forward,aiming at lowering the subjectivity and increasing the decision accuracy.The feedback mechanism and delegation process are incorporated,which can provide certain suggestions to the group and cope with the dynamic users throughout the decision process,respectively.Finally,the ultimate satisfactory solution can be obtained based on a collective group preference.This model can provide decision support for the management of weapon portfolio solutions.