Demand Forecasting for Pharmaceuticals
Open Access
- Author:
- Heininger, Sarah Elizabeth
- Area of Honors:
- Supply Chain and Information Systems
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Robert Alexander Novack, Thesis Supervisor
John C Spychalski, Thesis Honors Advisor - Keywords:
- Forecast
- Abstract:
- This thesis analyzes the demand an animal pharmaceutical company sees and compares forecasting methods to find the most accurate one. Companies collect large amounts of data and a forecast model can utilize the collected data to predict future demand. There are various forecasting techniques that are better suited for different industries. In this research, error terms identify which forecast technique a company should be using by pinpointing which method produces the forecast closest to the historical data. In order to create this analysis, eighteen months of data were provided by Company Z and three forecast models were set up in Excel. The Excel model relies heavily on data input and subject matter experts from the company to ensure that the correct data is utilized in creating the forecasts. To conclude, this thesis provides information on which forecast method was most accurate for Company Z and how they will continue to monitor and update the forecast to see what their demand will be for the next three to six months. The findings for the animal pharmaceutical company in this thesis identified that exponential smoothing is the best forecasting method to utilize. The major findings confirm that for a company that sees a relatively low amount of seasonality and year after year sees the same trends, exponential smoothing creates the most accurate demand forecast.