Proposição de um modelo de sorteio computacional adaptado à reserva de vagas para processos seletivos em instituições públicas

Abstract

Sweepstakes have been widely used as a mechanism to ensure fairness in the allocation of indivisible goods, such as tasks, housing, or seats in educational institutions. In the context of Federal Institutes in Brazil, especially after the COVID-19 pandemic in 2020, this practice became a recurring resource in the selection processes for student admission. However, conducting sweepstakes in a more transparent and auditable manner remains a relevant challenge for public institutions. The absence of standardization and clear criteria in the procedures, combined with social distrust toward computational sweepstakes and the complexity introduced by quota policies, highlights the need for a computational sweepstakes model that seeks to promote greater transparency and auditability while correctly identifying admitted candidates. Given this scenario, this research aimed to develop and evaluate three artifacts designed to improve the execution of computational sweepstakes, covering the allocation of seats in compliance with the reservation (quota) policy in public-sector selection processes. The artifacts include: (i) the BPMN modeling of the steps involved in conducting a sweepstakes; (ii) a guidance manual for carrying out these steps; and (iii) computational sweepstakes software. The artifacts were conceived using the Design Science Research (DSR) methodology, which guides the detailed construction and prescription of artifacts. The evaluation of the artifacts took place in two stages. First, through the presentation of the artifacts to key users involved in the IFRS selection processes, with the purpose of collecting perceptions and validating their applicability. Subsequently, statistical tests were conducted using data from a real sweepstakes, in order to technically analyze the computational randomness of the algorithm. The results obtained demonstrate both theoretical and practical contributions to the field of computational sweepstakes. The study deepens discussions on concepts such as randomness and auditability, while also incorporating software quality factors into the proposed solution. Moreover, the developed artifacts were structured to serve as a practical reference that can be adopted or adapted by other public institutions. Despite these advances, challenges remain regarding the feasibility of a full blockchain-based implementation, large-scale performance, interoperability with existing systems, and the need for broader institutional and social acceptance. [resumo fornecido pelo autor]