From task automation to knowledge automation : a systematic review of RPA limitations and prospective analysis of generative AI in this scenario

• Ano de defesa: 2025
• Autor(a) principal: Nunes, Izaque Dione
• Orientador(a): Thom, Lucinéia Heloisa
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Gerenciamento de processos de negócios; Automação Robótica de Processos; Inteligência artificial generativa; Automação inteligente
• Palavras-chave em Inglês: Task automation; Knowledge automation
• Link de acesso: http://hdl.handle.net/10183/298643

One of the tools used by Business Process Management (BPM) to optimize processes within an organization is Robotic Process Automation (RPA), which involves the development of software robots (bots). These bots are programmed to mimic human behavior with systems. RPA adoption in the global market is growing rapidly, but even with innovative investments and market enthusiasm, approximately 30% to 50% of initial RPA projects fail. The lack of a clear map detailing the most common barriers and offering a set of countermeasures, leads organizations to navigate a costly and inefficient process of trial and error. It is precisely this lack of systematized knowledge that constitutes the central problem investigated in this work. Therefore, the main objective of this work is to develop a taxonomy for RPA limitations in organizational settings, as well as to present the countermeasures defined in the literature, and to propose a prospective analysis of how the capabilities of Generative Artificial Intelligence can be applied as a new class of countermeasures to these limitations. To achieve this objective, a Systematic Literature Review (SLR) was conducted. The search was conducted in academic databases, resulting in a final selection of 94 primary studies for in-depth analysis. These were then subjected to a thematic analysis to synthesize the results and generate the taxonomy. As a result, 10 limitations were identified and cataloged, which were grouped into a taxonomy of three categories: Strategic, Organizational, and Technical Limitations. An analysis revealed that, in addition to technical challenges such as the fragility of robots, strategic and organizational barriers are critical factors for failure. The main contributions of this work are: (i) A consolidated taxonomy of RPA limitations, classifying 10 defined RPA limitations into strategic, organizational/human, and technical categories, providing a clear map of the problem space. (ii) A prospective analysis that maps Generative AI capabilities as solutions to the identified limitations, proposing a transition from "task automation" to "knowledge automation". (iii) A knowledge base and taxonomy for RPA limitations that can be used to develop an instrument to assess the maturity of RPA implementation in organizations. It concludes that Generative AI can be used to help overcome RPA limitations and has the potential to redefine the way process automation is performed. The synergy between the two technologies represents a fundamental transition from “task automation” to “knowledge automation”.