Neutrosophic-Logistic Hybrid Model to Evaluate the Effectiveness of Penitentiary Policies and Their Impact on Criminal Recidivism in Ecuador

Abstract

This study aims to develop a hybrid model that combines neutrosophic logic with logistic regression to evaluate the effectiveness of prison policies in Ecuador and their impact on criminal recidivism. The central problem lies in the inefficiency of the Ecuadorian prison system, characterized by high levels of intra-prison violence, deficiencies in rehabilitation and social reintegration programs, and a rising recidivism rate among inmates. This situation is exacerbated by the employment discrimination that ex-offenders face when reintegrating into society, which limits their opportunities and encourages the repetition of criminal behavior. The methodology applied was a mixed approach, with a non-experimental cross-sectional design and a descriptive-explanatory scope. Data were collected through surveys administered to inmates and former inmates, using neutrosophic scales to measure subjective perceptions such as the effectiveness of policies, the level of employment discrimination, and readiness for reintegration. These assessments, expressed as neutrosophic numbers (T, I, F), were transformed into indices that served as predictor variables in a logistic regression model, executed with statistical tools such as SPSS and Python. The results show that the index of perceived workplace discrimination and the low effectiveness of policies are factors significantly associated with criminal recidivism (p < 0.05), while participation in reintegration programs reduces this probability. It is concluded that the hybrid model allows integrating subjective perceptions with robust statistical analysis, providing a useful tool for decision-making in public policy. The findings support the need to redesign penitentiary strategies, focusing on social inclusion and effective rehabilitation to improve the safety and well-being of both inmates and Ecuadorian society.