Date of Award

2025

Document Type

Dissertation

Degree Name

Doctor of Business Administration (DBA)

Department

School of Business

First Advisor

Dr. Paul Shelton

Second Advisor

Dr. Brian Snider

Third Advisor

Dr. Dirk Barram

Abstract

This study evaluated the predictive performance of three forecasting models—Autoregressive Integrated Moving Average (ARIMA), Prophet, and Long Short-Term Memory (LSTM)—to determine which most accurately forecasts pediatric appointment availability. The research was guided by the theory that accurate forecasting can enhance healthcare scheduling by aligning appointment supply with patient demand, thereby advancing patient-centered care (PCC).

A quantitative, comparative design was employed using two years of pediatric online scheduling records. The methodological framework incorporated multiple forecast horizons (30–180 days) and statistical analyses, including ANOVA, paired sample t-tests, and regression. Model performance was assessed using standard error metrics: Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE).

Results indicated that ARIMA consistently outperformed Prophet and LSTM across all error metrics, validating its use in pediatric healthcare settings characterized by stable seasonal and linear patterns. Prophet showed strengths in capturing seasonal surges, while LSTM demonstrated potential for more complex datasets but was limited by the structure of the data and absence of exogenous variables.

The study contributes to forecasting literature and healthcare practice by demonstrating how predictive modeling can reduce scheduling inefficiencies, align capacity with demand, and strengthen PCC. Future research should explore integrating external variables and real-time data to enhance forecasting accuracy and care delivery further.

Download

Share

COinS