Simulating Forecasts Using Information Delays

Many business decisions use forecasts of future behavior (e.g., the sales rate for a product) based on past observations. In order to simulate such systems using GoldSim, it is necessary to simulate the forecasting process itself. For example, when modeling a supply chain, it is necessary to simulate how orders are placed. In the real world, these orders are placed not based on the current demand (which may not even be measurable), but on a forecast of future demand.

An Information Delay A delay element that delays information signals, and does not enforce conservation of the signal. It is intended to be used to simulate delays in measuring, reporting, and/or responding to information. can be used to represent a very commonly used forecasting method known as exponential smoothing. In exponential smoothing, the forecast is based on an exponentially-weighted average of past observations. Mathematically, this is equivalent to the output of an Information Delay with n =1. An example of such a simulated forecast is presented below:

When using an Information Delay to simulate an exponentially smoothed forecast, the Delay Time is a measure of how heavily weighted previous values are. The larger the Delay Time, the more heavily weighted older values are (i.e., the farther the forecast reaches back in time to generate the forecasted value). The smaller the Delay Time, the more heavily weighted more recent values are. A larger Delay Time will generally result in a smoother forecast, but one that is less responsive to the most recent values. A smaller Delay Time will generally result in a noisier forecast, but one that is very responsive to the most recent values.