Introduction As cancer has become the leading cause of death, the demand for services in oncology institutions has steadily increased in recent years. Some researchers have indicated that patients spend significantly more time waiting, both waiting to make an appointment and waiting at cancer institutions. Therefore, the article “Improving Patient Access to Chemotherapy Treatment at Duke Cancer Institute” (Woodall etc., 2013) aims to improve patient flow in their institution, while also focusing on the application in other oncology institutions. Initially, the authors obtain some basic information about the Duke Cancer Institute on the flow of information between different departments. Departments include clinic, radiology, central laboratory, oncology treatment center (OTC), and pharmacy; types of nurses include full-time and part-time. In order to optimize and simulate processes to meet patient demand and allocate resources, the researchers provide three models to achieve their goals. Since these three models are analyzed step by step and the next model is based on the previous results. The first model is the “discrete event simulation model”, which aims to predict patient waiting time and capture resource utilization information between different departments. However, researchers have identified that the most severe bottleneck is found in OTCs, as nurses are not available during the treatment process. Therefore the researchers decided to focus mainly on OTC and made assumptions about the maximum number of patients for each nurse. As a result, the counter service is subject to temporal variability, improving shift times and working hours of nurses will be the best way to distribute nursing supply with patient demand. Based on the first mode, they switched to using mixed integer programming model' for moving nurses, to eliminate the bottleneck in OTC. Shift schedules include daily, weekly and monthly, nurse type includes full-time and part-time. This method is used to focus on predetermining the number of nurses on weekly and monthly schedules. The full time nurse type will include 10 and 8 hour days. Weekly and monthly assignment to nurses is based on daily patient demand. As a result, these analysts decide to change one or two full-time nurses with the same skill levels as part-time nurses, which is better suited for peak demand over the counter and reduces overuse of resources when there are no there are too many patients. the final model builds on the previous model to further alleviate the bottleneck by optimizing the start and end time of nurses' daily shift. This method focuses on the daily movement of nurses.