A Decision Support System for Project Risk Management based on Ontology Learning

Abstract

Project Risk Management (PRM) is one of the main concerns of project management executives and professionals. Although PRM frameworks and risk models are mature enough to provide a systematic approach for managing risks, these practices remain ad hoc and non-standardized. In addition, there is no significant work shift toward PRM recommendation systems through inference rules and axioms. This study aims to bridge the cited gaps in PRM by developing a decision support framework based on an ontology that predicts personalized recommendations for managing PR processes effectively, and then making the right decisions. To this end, this framework takes advantage of the ontology semantic strengths to model a unified PRM knowledge relying on PMI’s framework. The idea is to parse PMI’s standard for PRM to enrich and exploit an existing PR Ontology. The enrichment process is driven by the Ontology learning (OL) tasks using Natural Language Processing techniques (NLP) to extract the main concepts, properties as well as OWL DL axioms and SWRL rules. Then, through Jena_rule engine, this decision system infers recommendations, by which a team member asks for a specific targeted risk-related request. Based on this approach, a decision system is developed to illustrate the assets of ontological reasoning and thereby the reliability of decision support. The potential benefits of the proposed framework are evaluated using a questionnaire survey that proves the overall positive evaluation.