Svirsko, Anna C
(2019)
Multiobjective and Robust Optimization in Pharmacy Delivery and Emergency Department Nurse Staffing.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Over the past 20 years, hospitals have seen a drastic improvement in patient service with improved patient recovery times, faster delivery of care, and an increased focus on patient safety. In addition, healthcare has also began to focus on its staffing, making efforts to improve staff satisfaction. Despite these advancements, healthcare spending has continued increasing. The objectives of cost, staff satisfaction, and patient safety do not always align. In addition, in the healthcare sector demand is frequently unknown, compounding the difficulty in solving these problems. This dissertation provides deterministic and robust optimization models to solve the pharmacy distribution problem and the emergency department nurse staffing problem.
Pharmacy distribution is one area of the healthcare system which has seen an increase in technology to improve patient safety and help facilitate distributing medication from the central pharmacy to the patient. Pharmacy robots are used to pick medication cost effectively in the central pharmacy while automated dispensing cabinets (ADCs) are point-of-use storage maintained on the inpatient units. These technologies have provided the ability to deliver medication more cost effectively and improve patient safety. However, the two technologies are often implemented independently of each other and therefore may not work cohesively within the distribution process. This dissertation presents a model which focuses on the trade-off between cost and patient safety. The model is solved to minimize the total distribution system cost, including purchasing and maintenance of technology, as well as minimizing the total distribution workload cost. A robust model is formulated and solved to account for the variation in medication demand and determine the effect variation has on planning decisions.
Emergency departments treat patients with various illnesses which results in a wide range of complexities in determining staffing. Patient demand varies by the day of the week and the hour of the day. This results in difficulty determining the daily staffing levels necessary to treat patients efficiently while also considering available staffing resources. This dissertation provides a basic model which determines the daily staffing levels for an emergency department based on predetermined staffing levels. This model is expanded by determining staffing levels in the presence of understaffing by assessing nurse to patient ratios. Several additional models are introduced which consider the variation in patient demand that occurs daily. The performance of these models is considered across a calendar year by comparing the results with current patient safety standards and in comparison with each other to determine the best method to determine nurse staffing levels in the presence of variable demand and understaffing.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
19 June 2019 |
| Date Type: |
Publication |
| Defense Date: |
18 March 2019 |
| Approval Date: |
19 June 2019 |
| Submission Date: |
22 March 2019 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
146 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Industrial Engineering |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
pharmacy delivery, pharmacy distribution, nurse scheduling, nurse staffing, robust optimization |
| Date Deposited: |
19 Jun 2019 15:08 |
| Last Modified: |
19 Jun 2019 15:08 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/36033 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}