This paper is concerned with the proctor assignment problem. Although there have been many proposed models for solving this kind of problem, it turns out that those models perform well only in their context. Also, no models can solve all problems in any context. Hence, creating a model for a specific problem is essential. In this paper, we define the problem and challenges of assigning proctors at our university as an optimization model. We experiment on two methods: the Generic Algorithm and the Integer Programming, then compare the results. For the Integer programming, we proposed an objective loss called zero-min loss, which minimizes the spans of days the proctors must be at the university during the examination time. Our main result shows that the proposed Integer Programming gives a 2285 higher score in fitness value compare to the Generic Algorithm.