Font Size: a A A

Research Of Resources Project Investment Risk Based On Bayesian Network

Posted on:2017-10-08Degree:MasterType:Thesis
Country:ChinaCandidate:K ChengFull Text:PDF
GTID:2359330488951583Subject:Management Science and Engineering
Abstract/Summary:PDF Full Text Request
With the economic globalization,market internationalization and the rapid development of science and technology,China has achieved rapid economic development,with the increasing overall size and improved development.But the development of China still dependents on the natural resources.Faced with a tighter resource constraints,the serious environmental pollution,the ecological degradation,we must set up the ecological civilization concept of respecting,complying with and protecting the natural,and choose the sustainable development road.Under such a situation,our country put forward the strategy of sustainable development and the principle of ecological civilization construction.Resources project refers to the natural resources investment projects,which relying on natural resources development.With the double pressure of internal and external,resources project investors face the threat of various uncertain factors,so it is necessary to study resources project investment risk management.Bayesian networks can be used in uncertain and incomplete information environment,basing on the causal relationship between each risk factor and have the ability of reasoning out resources project investment risk profile.This article is based on Bayesian network for resources project investment risk analysis research.First of all,risk management and Bayesian network aspects are reviewed in details.Then this paper introduces and reviews the resources project under the background of the current research situation and social actual situation.Then the basic theory and related characteristics of Bayesian networks are introduced.Based on the above theoretical basis,the history of literature,knowledge,the Delphi method and questionnaire method and so on,using many risk identification method of integrated application for resources project investment risk identification,and according to the risk level matrix,the influence degree of the risk factors for resources project investment and risk probability of comprehensive analysis,the fifteen risk factors in this study are chosen.On GeNIe software platform,firstly,using the expert knowledge for resources project investment risk factors of background knowledge in preliminary editing,then use GeNIe software ’s K2 algorithm data to do structure study,and combining the causality analysis further adjustment and optimization to constructs the Bayesian network structure;To set up a standardized processing of the data obtained from research on parameter learning of Bayesian networks,the conditional probability distribution of each node variable is obtained,and then to build resource-based Bayesian network inference analysis of project investment risk,through the analysis of the most broadly chain is the key risk factors,put forward targeted management advice to improve the resources project investment risk management.Finally,this article will build resources project investment risk Bayesian structure model applied in Jiangxi Yiyang county Y river channel sand mining projects.In Y county river channel sand mining project risk analysis,doing quantitative analysis of Bayesian network model,using risk resources project investment risk factors for state changes,through the model of the automatic update feature to update the entire network,understand their risk profile and calculate its investment performance expectations.Through the analysis of actual case,verify the feasibility of Bayesian network model.In this article,through the analysis of resources project investment risk identification and assessment,the Bayesian network modeling’s learning,inference and application,build and verify the Bayesian network model for resources project investment risk management.
Keywords/Search Tags:Resources project, Project investment, Risk management, Bayesian network
PDF Full Text Request
Related items