A hybrid method for evaluating the effectiveness of giant systems with indicator correlations: an application for naval formation decision making in multiple scenarios

This article discusses the giant system effectiveness evaluation (GSEE) problem with inevitable correlation in indicator systems due to their high specificity and complexity and proposes a hybrid method that is then applied to the naval formation decision-making process in multiple scenarios. The indicator correlation in a large-scale system will generate bias in its evaluation of effectiveness; the proposal that the lower the correlation is, the better the performance of the evidential reasoning approach (ERA) has been proven mathematically. In light of this proposition, a corollary was put forward: Fewer indicators would improve the precision of the result of the ERA application when considering the correlation. Considering that the giant system can be split into respective subsystems, which can then be analyzed by experts in their own fields, a hybrid method was developed for the GSEE problem based on the ERA and prospect theory. The core of the method is the construction of a nonlinear optimization model (NOM) aimed at minimizing the correlation and maximizing the evaluation ability of the prospect value of the indicator system. By constraint, the NOM also includes the optimized weight value of each indicator. For demonstration purposes, a naval formation operation effectiveness evaluation (NFOEE) was performed to assess the feasibility of the proposed method and the NOM. The results show that the proposed method can solve the NFOEE effectively and allow the decision maker to obtain useful information for naval formation-type decisions in multiple scenarios. Furthermore, the evaluation method is a general tool that can be applied to other GSEE problems.