Resumo:
Brazilian public works are generally viewed negatively by the population, this image is mainly linked to construction stoppages, delays, extrapolation of costs and poor quality. Efficient project management has favored public organizations in improving the performance of construction management. This dissertation presents a proposal to prioritize the implementation best practices of project management practices as a way to improve the management processes of the works of a federal public university. The research used the Soft System Methodology (SSM) as the main methodology, supported by the application of a Systematic Literature Review (SLR) and the GUT Matrix to prioritize good practices. A diagnosis of the main problems of the institution that are related to the management of the works was carried out. Aided by SLR, a conceptual model was built, containing 42 (forty-two) good practices applicable in public works, which can be used as a reference in future research on the subject. This model was validated by experts in the field through a questionnaire, which pointed out the most important good practices for the investigated context. Finally, the GUT Matrix tool was applied, which used the criteria of Severity, Urgency and Tendency, based on the evaluation of professionals who work directly with the builds, to propose good practices as priorities: (1st) “qualitative risk analysis” , (2nd) “requirements analysis”, (3rd) “contingency plan”, (4th) “stakeholder analysis”; (5th) “quality plan”; (6th) “analysis of the critical path method”; (7th) “project management models”; (8th) “lessons learned; (9th) “progress meetings”; (10th) “Project Management Software” and (11th) “Business Information Modeling”. Based on the order of prioritization presented in this dissertation, the administration of the institution will have greater basis to make decisions that aim to improve the management of the works carried out on the campus, aiming to reduce delays in the delivery of works and to have greater efficiency in the work performed by the servers.
