Research On The Risk Measurement Of Project Portfolio Influenced By The Interaction Effects

In the economic environment of diversified market demands and increasingly fierce competition, the trend of projectification becomes more and more prevalent in enterprise. Moreover, driven by the New Product Development and Information Technology, the challenge of parallel operating multiple projects and maximizing the economic benefits based on the complementary advantages between projects has been faced by many enterprises. Therefore, since 1980 s, project portfolio management has become a hotspot in the field of project management, which leads the research of project portfolio risk, one of the key issues of project portfolio management, to be having theoretical value and management practice significance. However, so far, the research of project portfolio management focuses on the portfolio selection rather than the risk of project portfolio. What’s worse, within the current study of project portfolio risk, the impact of the interaction effects on the risk of project portfolio has not yet been enough considered and the risk formation mechanism and risk constitution has not been deeply analyzed. Moreover, ignoring the differences between project portfolio and capital portfolio, most researches associated with the measurement of project portfolio risk focuse on the methodology in financial category. Thus, it is urgently needed to consummate the research of the project portfolio risk, which includes considering the impact of the interaction effects on the portfolio risk as well as constructing the risk measurement model being consistent with project characteristics.This research aims to systematically analyze the impact of interaction effects on project portfolio risk starting from analyzing the complexity features of the portfolio risk, which includes revealing the influence of interaction effects on different levels of the formation of portfolio risk, identifying the risk constitution and measuring it through constructing the measurement frame and multi-dimensional metric models of the portfolio risk. On the whole, this research analyzes the resource, features, and the measurement method of the project portfolio risk influenced by the interaction effects from qualitative to quantitative. The main research contents and results of this thesis are listed as follows.(1) The nonlinear impacts of different scales on the project portfolio are analyzed through quantifying the correlated attributes between projects, which reveal the substansive characteristics of the portfolio risk complexity. Then, the system evolution process of the portfolio risk from the single project risk is presented based on the aforementioned research, which is consists of three stages: the nonlinear evolution of the accumulation of projects, the risk evolution of the system structure, and the risk evolution of the complex management environment.(2) The comprehensive measurement frame of the project portfolio risk is constructed based on the different impacts of interaction effects on the different phase of the system evolution process of the portfolio risk, which can realize the comprehensive measurement combined the objective measurement from various perspectives and the subjective risk preference of managers. Through analyzing the differences between evolution levels, the source of uncertainty in the portfolio are identified on the basis of the system evolution of the project portfolio risk. Thus, the portfolio risk is to be including the accumulation risk formulated by the nonlinear evolution of the projects accumulation, the brittleness risk formulated by the risk evolution of the system structure as well as the conflict risk formulated by the risk evolution of complex management environment. Then, according to the different impact of interaction effects on the portfolio risk whose competition results in the portfolio risk, the risk is divided into the risk of positive utility characterized by the risk diversification and the risk of negative utility characterized by the new risk. To be specific, the positive utility risk is the accumulation risk which has the posibilty to be smaller than the aggregated risk of projects implemented seperately and the other one is consists of the brittleness risk and conflict risk which are the new risks emerge only in the context of the project portfolio. Finally, combined with the subjectivity of risk, the competing measurement frame is constructed from the perspective of the positive and negative impact of the interaction effects, which can reflect directly and measure the power contrast status of different utility risks and the overall risk change influenced by the risk preference of managers on the basis of the objective measurement of multi-dimensional risks.(3) The appropriate measurement methodologies are selected according to the risk properties in different evolution levels and the measurement results are analyzed contrastively using the same example based on the competing power model of project portfolio risk. For the accumulation risk of the project portfolio, as the Bayesian Network has a powerful capacity to express and inference the uncertainty knowledge, the network which express the nonlinear dependency relationship between project risks is built based on the Bayesian Network. In the process of constructing the network, the magnitude of correlation between projects is measured using mutual information and the group search algorithm which can improve the efficiency of modeling is presented to traverse nodes in the network. Finally, the risk diversification capacity of the project portfolio is calculated based on the accumulation risk which is measured using the probability inference of the Bayesian Network.(4) The brittleness risk measurement model is constructed based on the brittleness theory and modeling method of complex systems. The brittleness risk of the project portfolio is analyzed according to the brittleness theory. Then, based on the brittle link entropy and brittleness risk entropy, the brittleness risk degree is presented to assess the overall uncertainty of the whole failure of the portfolio. Furthermore, in view of the seriousness of the brittle hazard, based on the graph theory, the network model of the brittleness risk spreading is constructed by measuring the brittle link probability, by which the brittle source and brittle collapse path are identified and the risk of the overall failure of the portfolio results from the brittle source is calculated.(5) The conflict risk measurement model is constructed based on the Fuzzy Comprehensive Evaluation. The conflict risk of resource allocation and coordinate implement are inferred to arise from the complexity of the management environment caused by the interaction effects through analying the management environment of the project portfolio, including each stage of the project portfolio management, management roles, and the interplay between different management roles. The influence factors involved in the occurrence and resolution of the conflict risk are determined to comprise the complexity of the interaction between projects and eight manager competency factors which identified by the risk checklist made by using the manager competency standard. Finally, the multi-dimensional Fuzzy Comprehensive Evaluation model is established, which uses the synthetic weight combining the objective weight reckoned by the method of entropy weight and the subjective weight calculated by the Analytic Hierarchy Process.