Multi Criteria Decision Making Techniques and Knapsack Approach for Clustering, Evaluating and Selecting Projects

Project Portfolio Selection is one of the strategic decisions on the level of management in project-oriented organizations; which is one of the most important and effective stages in project portfolio management. In other words, after identification and evaluation of different projects, optimal combination of projects must be selected based on different criteria. Since commonly resources of organizations are less than required resources on projects that are ahead of the organization; so project selection among different project portfolio and decision making about these issues are inevitable in organizations. So durability of organizations depends closely on the way of modeling and used approaches for project selection problem. In previous researches related to project selection problem, generally the problem of project clustering have not deeply been considered while project clustering may allow that projects to reach the highest return. In this article, we will propose a model for clustering, evaluating and selecting projects. At first projects will be clustered in various portfolios by K-MEANS algorithm, then projects of each portfolio will be evaluated and prioritized with analytical network process (ANP). Finally, projects from each portfolio will be selected based on knapsack problem.