Delayed graduation is a factor of concern at the master's level. There is some information available in the literature on the subject worldwide yet centered on undergraduate programs. However, there is limited research on students at the postgraduate level. Researchers from all universities in Chile and worldwide have discussed delayed graduation as it is a critical consideration for all institutional accreditation processes. For this reason, educators and researchers in institutions are interested in offering strategies to reduce the graduation time by analyzing risk factors students face during their training. This study presents an early detection model based on machine learning to account for graduation delays seen at the master's student graduation process. This article presents a descriptive study that examines the relationship between students' characteristics (gender, age, education program, and qualifications) and program characteristics (program duration, program location, and class schedule) with delayed graduation. The analysis takes a sample of students (1257 records) from the last five years who have completed all program credits for their engineering master's degree and have registered in the final project process. Using descriptive analysis, the study documents the factors that play a crucial role in explaining delayed graduation. The risk factor database includes individual student information such as age, gender, and academic information such as course passing rate, faculty, dropout rate, and graduation time. Study results reveal that students who fail the first subject in their master's thesis tend to increase their graduation time considerably. We found factors affecting delayed graduation, including individual characteristics (gender; with low weight-significance), students' backgrounds (education, pass rate; with high weight-significance), and institutional environment (program type/modality of the program, location; with medium weight-significance). Results indicate that it is necessary to support students to improve their performance. Higher education institutions should use this study's results to develop proactive initiatives (student accompaniment and academic staff accompaniment) that might help reduce students' graduation time.