多偏好等級工作班次與休假日之護理師班表排程研究

Abstract

為護理師規劃一個公平且滿意的班表，是近年來護理師班表排程中重要的議題。過去研究中護理師的班表滿意度，只著重目前規劃週期內全體人員偏好的工作班次與休假日之總數量。然而，這種滿意度的設計有一些缺陷，如護理師對不同的工作班次或休假日有不同的偏好等級，隨著過去這些等級被滿足的程度會影響班表的公平性；及因工作班次與休假日的偏好數量是不相等以導致班表結果有偏差。因此，本研究考量工作班次與休假日的偏好權重、多偏好等級的工作班次與休假日、指派人員的優先順序及過去資料，提出一個新穎的滿意度公式。此滿意度公式能公平地滿足護理師偏好的工作班次與休假日，使護理師的班表總滿意度最大。首先，本研究考量排程限制，建構一個整數規劃模型去求解小規模問題中班表的最佳解，接著發展三個改良的基因演算法包括改良型基因演算法、混合型基因演算法及移民基因演算法，去搜尋較大規模問題中班表的近似最佳解。最後，基於高雄市某醫院的門診資料做實驗模擬，實驗結果顯示，班表的規劃結果中護理師偏好的工作班次與休假日的滿足程度達到80%以上。在研究的貢獻方面，護理師的班表滿意度會受到偏好等級及優先順序之影響，在班表品質的衡量上更合理。

It has been important to make a fair and satisfactory schedule for nursing staff. In the previous works, the satisfaction of nursing staff for schedule was usually based on the total amount of assignments of the preferred shifts and days-off of the nursing staff at the current planning period. However, the design of such satisfaction has some flaws, as nursing staff have other different preference ranks for shifts and days-off, which may affect fairness of schedule, and numbers of the preferred shifts and days-off are not equivalent so that the planned schedule might be biased. Therefore, this dissertation proposes a novel satisfaction which takes into account the balance of the preference weights for shifts and days-off, different preference ranks towards each shift, the priority ordering of the nursing staff for planning their shift schedule, and the historical data of previous planning period. The proposed satisfaction can fairly satisfy all the nursing staff’s preferences for shifts and days-off to maximize the satisfaction of the nursing staff for their schedules. In addition, the dissertation considered schedule constraints to build a model based upon integer programming to an optimal solution of schedule in small-scale problems, and develops three improved genetic algorithms including improved genetic algorithm (IGA), hybrid genetic algorithm (HGA) and genetic algorithm with immigrant scheme (GAIS), to search near optimal solutions of schedules in large-scale problems. The main contribution of our research is that we consider that the nursing staff’s satisfaction is affected by multiple preference ranks and their priority ordering to be scheduled, so that the quality of the generated schedule is more reasonable.