The Repository @ St. Cloud State

Open Access Knowledge and Scholarship

Date of Award

8-2013

Culminating Project Type

Thesis

Degree Name

Applied Statistics: M.S.

Department

Department of Mathematics and Statistics

College

College of Science and Engineering

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

The aim of this research is to forecast patient volumes in the Emergency Department of a regional hospital in Minnesota, which eventually will aid in addressing the issue of registered nurse staffing fluctuation, more specifically, productivity and capacity planning in the ED. Several methods are applied to forecast arrival patient volume, and cumulative patient volume to evaluate each model’s performance. The methods considered are linear regression, time series models and dynamic latent factor method. Long term forecast for as long as six months ahead is the goal here due union regulations that only allows for significant changes in registered nurse staffing schedule be put in place six months in advance. This long term forecast will enable administrators implement effective and timely changes to enhance productivity.

The patient arrival count, where each patient is counted once in the system, is analyzed to see how many patients the department encounters hourly. Also, cumulative patient count which gives us an idea of how many patients are in the department at any given time was also considered, here patients are counted for every hour they are in the emergency department (ED). Patient who come to the ED are categorized by their acuity level. Of all the patients that came to the ED, 52% need urgent care; this group is also analyzed to predict their arrival volume.

Lastly data was simulated with different patterns and the forecasting results from the different methods were compared and estimated. The forecast accuracy and performance for these models is then evaluated using out-of-sample forecasts for up to six months ahead. Mean square error (MSE), Root mean square error (RMSE) and mean absolute error (MAE) were utilized tosee which method is most reliable and also consistent.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.