| Summary: | 碩士 === 國立屏東科技大學 === 工業管理系 === 91 === The assignment of nurses is one of the important managerial tasks in health instituties. It affects not only the cost of human resource but also the environment where nurses work at. These facts can all influence the quality and efficiency of nurses’ work. In fact, all the nurses in hospitals use three-shift working schedules to take care of patients around the clock. Additionally, all nurse rosters have to obey the laws and regulations set by the government, and follow the special dictates made by the hospital and labor union. This is called “Hard Constraints.” There are also some special conditions that are made to satisfy nurses’ preferences called “Soft Constraints.” The optimal nurse rostering problems are commonly modeled by goal-programming models with binary decision varialbes. Due to the combinatorial characteristics, optimal nurse rostering problems are classified as NP-complete and are difficult to solve when their sizes become large.
This research presents a goal programming model to solve an optimal nurse roster problem. The hand constraints encompassed in the model are: to meet the daily demand of nurses and to obey the Labor Standards Law. The model also includes the soft constraints, which are to avoid 010, to work one shift per day, to work on the same shift over one period, to have at most two consecutive days off among 4 days off in one period, and to assure the final roster only cause very little inconvience for all nurses. The objective function is to safisfy a pre-selected schedule and the soft constraints as much as possible, subject to the hard constraints. The siummlated annealing algorithm is used to solve the proposed model. The design of experiments (DOE) is used analyze the parameters associated with the simulated annealing. The results show that the model provides multiple solutions of good quality. It can also provide nurses and hospital employee a standard to follow.
|
|---|
