Chapter 3. Workload Forecasting

Forecasting takes information available in the present and uses it to predict the future . Everyone forecasts things. We use our knowledge of the past to decide what time to get up in the morning, how much food to buy at the grocery, and where to go on vacation. In daily life we don’t usually think about how we make these decisions or how we evaluate the outcome of our choices. Business forecasting is more structured. There are specific techniques to model the relationships between present information and a future value.

Few forecasts are exactly right. Some error is expected—some difference between the forecast and the actual value. In most cases, then, it is not enough to make a forecast. We also need to know how accurate the forecast will be. This means that the prediction is not an exact value but a range with a known probability.

This chapter looks at predicting the workload for a typical business process. The example forecasts the volume of calls coming into a call center, but the techniques can be applied to many other problems. This type of forecasting uses only the past values of the item being predicted. Future call volumes are predicted using past call volumes.

This chapter also demonstrates techniques for creating a complete application in Excel. The application uses a combination of organization, workbook functions, formatting, and a little VBA. At the end of the chapter we will have a complete Excel application for predicting workload. ...

Get Analyzing Business Data with Excel now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.