A Study On Allocating Nursing Resources In Intensive Care Units

ICU is an indispensable medical unit to modern hospitals, and it's also the last line of defense of life in critically ill patients. As the cornerstone of ICU's normal operation, nurse is an extremely important medical resource. The shortage of trained nurses is a worldwide problem, which is particularly severe in China. Good shift pattern and nurse schedules can not only reduce the costs of hospital's daily operation, but can also improve nursing care quality, alleviate the work pressure and enhance the job satisfaction of the nurses. Therefore, how to make good scheduling decisions and allocation of nursing resources with confirmed facilities, have caught great attention of hospital management.This thesis focuses on nurse scheduling/rostering problems in Chinese ICUs. Using mathematical models and heuristic methods, the research studies the advantages and disadvantages of different shift patterns and explores new self-scheduling method suitable for Chinese hospitals, whose performance is evaluated and analyzed comprehensively later. The main work of this thesis is as follows:(1) Analysis and comparison of three common shift patterns in Chinese ICUs (traditional pattern, three-shift pattern and two-shift pattern) is presented in this part. Several 0-1 integer programming models are established and similar heuristic scheduling methods which consider fairness are designed. Through cases based on same workload, the proposed method is verified. Suggestions on the selection of different patterns are provided from aspects of costs, workload and fairness.(2) Considering the flexible scheduling method in Chinese ICUs, a new method of self-scheduling is proposed in this part. Corresponding 0-1 integer programming model is established and a two-stage heuristic scheduling rule is designed based on weights of nurses'preferences and fairness. Furthermore, a tabu search method is provided, whose goal is to enhance the fairness of scheduling. The experimental results show that the tabu search method in this paper is stable and universal, and can better satisfy the preferences of nurses.(3) Further research related to the self-scheduling method proposed in this thesis is done in this part. Based on three cases with different preferences of nurses and using the two-stage heuristic rule and tabu search optimization method, the new scheduling method is compared with three other methods belong to three-shift pattern for obtaining the influence of nurses'preferences and flexible shifts to scheduling results. The study shows that although the self-scheduling method considering flexible shifts is slightly unfairness than other methods, it has obvious advantages in vacation and/or meeting nurses' preferences.