Abstract

Izabela NIELSEN, Peter NIELSEN, Quang-Vinh DANG

Abstract : When planning the best execution of a portfolio of projects, decision makers face the problem of making an optimal decision that meets an organization’s objectives and priorities in uncertain situations, under a given set of constraints with various sources of knowledge. This problem is typically reduced to schedule operations in projects in a cost optimal manner on shared resources with a given set of deadlines and constraints. This is in some instances an oversimplification of the managerial problem being addressed. In some cases, the complexity of the problem implies a necessity to decompose it into sub-problems and solving these individually before enabling an evaluation of the combined solution. This combined with a need to incorporate experts’ knowledge facilitates formulating the problem as a Fuzzy Constraint Satisfaction Problem (FCSP). Our goal in the context is to provide a Constraint Problem (CP) based approach facilitating both fast generation of project portfolio variants and a multi-criteria evaluation on the basis of expert’s knowledge. To achieve this goal, four distinct contributions are presented: a model based on a combined FCSP and CP approach, a new methodology for generating variants of plans and project portfolio evaluation, a new approach for fuzzy multi-criteria project evaluation taking into account company specific criteria and a Decision Support System (DSS) for project driven decision-making based on constraint satisfaction.

Keywords : Project management and scheduling, multi-criteria evaluation, decision support systems, constraint programming

Year : 2012
Issue : 2
Volume : 4

Not Available