| This dissertation presents an interactive knowledge-based decision model to evaluate project portfolios, to select the optimal portfolio, and to allocate resources to it. The model integrates and extends the attractive features of scoring techniques, analytic hierarchy process (AHP), and 0-1 goal programming. The approach consists of two phases: phase I--the screening phase, and phase II--the selection and allocation phase. During phase I, which is done outside of the interactive model, the scoring approach is used to eliminate unattractive projects. In this phase, the strategic fit of the projects are examined. Projects that are not congruent with the mission and objectives of the organization as well as those which do not meet certain minimum requirements set by the decision maker(s) are eliminated. During phase II, the decision maker, in the English language, provides the system with the data about the project characteristics and organizational requirements. Translating the English language into absolute constraints, goals, and nonabsolute constraints, DSS develops a 0-1 goal programming model. To prioritize the goals, the output of AHP or the direct instruction of the decision maker is used. Solving the resultant 0-1 goal programming problem, the model evaluates the feasible portfolios based on various criteria such as: (1) Economic attractiveness of the projects measured by their expected net present value; (2) Uncertainties of the technical and commercial success as they impact the benefits of the projects; (3) Availability of the resources required for each project; (4) Managers' perception of a balanced portfolio as it relates to basic and applied research projects, offensive and defensive projects, in-house and contracted projects, etc.; (5) Subjective considerations of the R&D staff and marketing personnel about the success and failure of each project; (6) Mutual exclusivity of projects; (7) Existence of one or more types of interdependencies (cost, benefit, and/or technical interdependencies) among projects; (8) Relative attractiveness of ongoing projects in comparison to new proposals.; The model has the the capability to address the nonlinear relationships between the amount of resources allocated to the project and its technical success rate. Also, the model incorporates the time variant characteristics of the parameters and priorities into the final decision. (Abstract shortened with permission of author.)... |
