| Project Portfolio Selection Problem (PPSP) is to select a set of projects from candidate projects under certain resources constraints in order to realize enterprise's one or more strategic objectives. PPSP is a sort of synthetical problem which involves operational research and decision-making technology such as combinatorial optimization, constrained optimization, multi-objective optimization, etc. PPSP has been widely used in many fields, such as investment project portfolio selection, IT project portfolio selection, Research and Development(R&D) project portfolio selection, etc. With the expansion of the PPSP, intelligent optimization algorithm is then adopted to solving the problem. PPSP has important theoretical and practical value.With the development of society, PPSP has developed on the basis of classical project management optimization problem. The classical project management optimization problem includes time-cost trade-off problem,Resource-constrained Project Scheduling Problem (RCPSP), robust resource-constrained project scheduling, etc. These problems were often researched in single project environment. Only in recent years, they have expanded to multi-project area, such as multi-project scheduling. Generally speaking, researches of these optimization problems only consider how to do a good job in projects, but not considering do good projects from enterprise's strategic height. PPSP considers both enterprise's macro and micro factors, it is more complex and has more vitality, so it is becoming one of hot topics in project management area.On the basis of a comprehensive overview on the research of project management at home and abroad, the PPSP is researched in the macro context. The modee of operation optimization is selected to set up the research model.PPSP is NP-Hard, therefore intelligent optimization algorithm is adopted to solve the constructed model. For solving the more complex models, Path Relinking (PR) is adopted to hill-climbing search, which ensures more accurate solutions. Simulation results show the effectiveness of the proposed algorithm. Thought systeming of characteristics of project portfolio in several fields, Combining with related literatures of home and abroad and characteristics of projects, we get the main breakthrough point, and the basic related theoretical knowledge about problem were collected and carded.The main achievements of this dissertation are as follows:(1) Project Portfolio Selection Problem considering Resources Conversion was researched. Resources conversion theoretical framework was proposed. Further, the concepts of constraint false violation and constraint true violation were put forward. Simulation test show, with considering resources conversion condition, resources can be integrated and utilized sufficiently, and more benefit of project portfolio selection be obtained,(2) Project Portfolio Selection Problem considering skill synthesis about multi-skilled workforce was researched. A function of skill nonlinear synthesis about multi-skilled workforce was constructed. A nonlinear mixed integer programming model was established, and the objective was maximum of the value of projects, which has been selected with constraint of numbers of multi-skilled workforce. Work team of multi-skilled workforce was built up for each project that has been selected, with considering skilled requirement of the project. Because nonlinear synthesis of skill, a three stages algorithm was designed to solve the problem. Simulation test showed the algorithm was effective.(3) Project Portfolio Selection Problem considering two-stage and double scenarios was researched under the whole portfolio constrction. A model of bi-level0-1integer programming model was established. Upper-level model was in deterministic scenario, optimization objective was maximization all of the value of projects, which have been selected within two stages. Lower layer model was in uncertain sense, optimization objective was maximization of expected value of projects that have been selected at the second stage, and the ratio of variance and expected value was adopted as risk constraint. Through test analysis, we get the optimization result with different risk index, and some robust core projects were presented.(4) Project Portfolio Selection Problem considering Multi-phase Rolling portfolio constrction was researched under uncertain condition. The optimization objective was maximization of benefits of each phase. Related constraints were builded relying both matching of strategy equilibrium between enterprise strategy requirement and synthesis of strategy contribution of projects which were selected, and the enhancement effect of resources. With this,we built a optimization model. Simulation test of ten phases show that enterprise can get benefits steadily during five phases, and benefits is differentiate about6-8phases. This conclusion can help enterprise adjust its strategy periods. Statistics about numbers of project are selected in every phase show the periodic regularity. Research resultion can give a guidance to execution of project portfolio selection.(5) Two models of Robust Optimization of project portfolio selection with fuzzy scenario was proposed, which objective function was Min-Max Regret or Min-Max Relative Regret. Combining risk preference of decision-maker, four deterministic modes about the models were formed. At the same time, the balance of the robust solutions feasibility and optimality was considered. Simulation test get lots of favorable conclusion, and some useful management advices are proposed.This paper has two main innovations, the first is the innovation of optimization constraint form, including the innovation of resources conversion constraint, multi-skilled workforce constraint, fuzzy scenario constraint, and strategy equilibrium constraint.the second is the innovation of project portfolio structure, including the innovation of Multi-phase whole project portfolio structure and Multi-phase rolling project portfolio structure. The main research achievements of this paper are: Resources Conversion mechanism and discriminant mechanism of constraint violation are discussed and applied to PPSP under uncertainty, the theoretical system of matching of strategy equilibrium is constructed based on fuzzy theory and relative formulas are given, matching of strategy equilibrium and enhancement effect of resources are used as association constraint. the concept of nearness in fuzzy theory is extended, the concepts of Upper-side and down-side nearness are proposed which enriches the contents of fuzzy theory. Some positive achievements are also attained through simulations, for example, optimum adjustment periods of enterprise strategy alteration.; periodic regularity between project arriving and project selection; the effect of personnel skill for project portfolio results and personnel assignment,etc.
|
PDF Full Text Request