Percepção de valor dos Fatores Críticos de Sucesso do projeto (FCS) sob a visão do cliente: uma proposta de sistema de inteligência artificial baseado em redes neurais artificiais para priorização dos FCS

Detalhes bibliográficos
Ano de defesa: 2023
Autor(a) principal: Sakalauskas, Eduardo de Carvalho lattes
Orientador(a): Bizarrias, Flavio Santino lattes
Banca de defesa: Bizarrias, Flavio Santino lattes, Scafuto, Isabel Cristina lattes, Martens, Cristina Dai Prá lattes, Machado, Michel Mott lattes, Martens, Mauro Luiz lattes
Tipo de documento: Tese
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Nove de Julho
Programa de Pós-Graduação: Programa de Pós-Graduação em Gestão de Projetos
Departamento: Administração
País: Brasil
Palavras-chave em Português:
Palavras-chave em Inglês:
Área do conhecimento CNPq:
Link de acesso: http://bibliotecatede.uninove.br/handle/tede/3160
Resumo: The objective of this thesis is to propose an artificial intelligence system based on artificial neural networks for prioritizing the critical success factors of the project (CSF) through the demographic profile of the client. Understanding the priority of the FCS from the customer's point of view can help the project manager to focus resources on items that can satisfy the customer, thus improving the project's chances of success, since the literature presents customer satisfaction as one of the main factors. items for achieving project success. This thesis was divided into three related studies that seek to: i) identify the direction of the CSF indicated in theory, ii) identify the difference in the perception of project success between the project manager and the project client, iii) identify the CSF that represents greater value for the project's client, iv) relate the client's profile with the FCS and v) create an artificial intelligence system based on neural networks that prioritize the FCS. Through paired bibliometrics with research from the last 10 years on FCS and using Exploratory Factor Analysis (EFA) as a statistical technique, it was possible to identify the academic direction of the theme and that discussions on FCS touch on five main areas, namely, i) Public-Private Partnerships (PPP), ii) Sustainable Projects, iii) Software Development, iv) Construction and v) Public Projects. It was also evident that the FCS is elucidated from the perception of specialists, teams, or GP, and even though they have in common customer satisfaction as a fundamental antecedent for the success of the project, there are no studies in which the client demonstrates which FCS is most important, it was also possible to create a single list with common FCS across projects. In the second study, a satisfaction survey was carried out with clients of 84 projects considered successful in the context of the Brazilian industry, showing that a percentage of clients were not satisfied and that the GP's perception may be biased. The analysis of the answers indicated that the GP has a more optimistic view of the project and that the client did not have the same perception about the project, that is, while for the client the project did not indicate complete success, the perception of the GP was the opposite and that the project had been a success. For the third study, a survey was carried out with 347 project clients, mostly production leaders, managers, and directors working in the Brazilian industry and responsible for project validation in their respective areas, in which it was possible to relate FCS to the profile of the client. This relationship was achieved through an artificial intelligence system known as machine learning based on artificial neural networks, this method simulates the functioning of a human brain, and in addition to allowing the system to learn from new data inserted in the response base, it also it is possible for the parameters to be altered, modified or even increased, adapting to the context of any project-based company, that is, if the client's behavior changes over time, the system will learn and modify the indicated prioritization, or if new surveys identify relationships with other profile data, the system will also receive these parameters and consider them in the response, and even if the FCS change over time and by type of project, the neural network can adapt and relearn with the new data.