Risk Analysis And Decision Aid For Building Overseas Strategic Base Under Incomplete And Sparse Information Environment

The opening of the "overseas strategic base"(OSB)will help to "accelerate the construction of a maritime power".Additionally,OSB plays a role in guaranting the safety of China's maritime energy channel transportation and seeking comprehensive maritime rights and interests.However,opportunities are often accompanied by risks.Under the complex background of international insecure politics and military,the construction of OSB requires comprehensive consideration of many uncertainties,such as the harsh natural environment,unstable geopolitics,different religious culture and limited observations,just to name a few.Therefore,the risk assessment of building OSB is rigorously meaningful.In order to complete the risk assessment tasks robustly and rationally,the main research conducted in this paper are as follows:(1)Risk identification of OSB has been implemented with due care,including the analysis of various risk factors.Thus,we construct a multi-level risk indicator system,which takes both natural hazards and cultural risks into consideration;(2)This paper summarized the physical charateristics of several natural hazards such as strong winds,huge waves and low visibility.Then we proposed a new way(so-called causality analysis)to detect the causal relation between the tropical cyclone(TC)genesis over the western North Pacific(WNP)and a variety of climate modes.With the selected climate indices as predictors,a method of fuzzy graph evolved from a nonparametric Bayesian process(BNP-FG),which is capable of handling situations with insufficient samples,is employed to perform a seasonal TC forecast.We found that the BNP-FG model together with the causality analysis can provide a satisfactory estimation of the numbers of TC genesis observed in recent years.;(3)Aiming at the fragmentation and inconsistency of data on natural hazards,we propose a new fast density clustering by information diffusion(RLCA-IDM).We can automatically detect information sources from all data points as the cluster centers using RLCA-IDM.Then,each point is assigned to the cluster of its nearest clustering center with more information content.The border region from the classic algorithms is employed and optimized to identify noise.Finally,we designed and conducted a series of experiments to demonstrate the newly proposed algorithm.Several state-of-the-art variants of RLCA are also included for comparison.The clustering results of both synthetic and real-world datasets indicate that the new clustering algorithm can be an effective approach for most datasets;(4)Considering the complexity of the cultural risk indicators for OSB,we propose a new variant of kernel entropy component analysis(KECA),namely,L1-norm-based KECA(L1-KECA),for data transformation and feature extraction.Accordingly,we present a greedy iterative algorithm that has much faster convergence than conventional one.Additionally,L1-KECA retains the capability to obtain accurate density estimation with very few features(just one or two).Extensive experiments on different real-world datasets validate that our model is superior to most existing KECA-based approaches.The code has also been made publicly available;(5)In order to provide a list of many host countries sorted by the ascending order of risk assessment,we propose a new clustering algorithm called the ordered fuzzy c-means clustering algorithm(OFCM).Different from the classical fuzzy c-means clustering algorithm,we use the net outranking flow of PROMETHEE and validity measures for clustering to establish a new objective function,whose properties are mathematically justified as well.Finally,we employ OFCM to solve a practical ordered clustering problem.A comparison analysis with existing approaches is also conducted to validate the efficiency of OFCM;(6)To objectively handle the multi-criteria group decision making(MCGDM)problems in which there is hesitancy in providing linguistic assessments and incomplete linguistic information of OSB,we firstly optimize the comparison method of probabilistic linguistic term set(PLTS).Then we propose a new ordered weighted averaging operators for assigning weights.A new fuzzy set called interval-valued probabilistic linguistic term set(IVPLTS)is also developed.The associated new operations,comparison laws and aggregation operators are designed for IVPLTS.Furthermore,we establish an efficient framework for MCGDM problems based on the proposed comparison method and the fuzzy preference relation.We apply it to a real-life case concerning OSB under linguistic environment.The extended TOPSIS methods combined with PLTSs by using different operational laws are also included for comparison.The final results demonstrate the efficiency and practicality of the new framework.This paper is expected to be applicable in building overseas bases for Chinese military.