Lesson 4: Building and Running a Simple GoldSim Model

Time to complete Lesson 4: 60 minutes

Are you ready to build a model? Let's build a copy of the model we just explored, one step at a time. To begin, we'll restate the problem:

This simple model simulates a pond of water that is used for forest fire suppression. A stream flows into the pond.The inflow rate varies from day to day, and this variation can be described statistically. There are no streams leaving the pond, but water seeps out of the bottom of the pond (so if the inflow stopped, the pond would eventually drain completely).

Over the next 100 days, the pond will need to be used for fire suppression. Fires are assumed to occur approximately once every 50 days. Whenever a fire occurs, a certain amount of water (100 m³) needs to be immediately removed (e.g., via helicopters) to fight the fire.

We want to know the likelihood that the pond will be emptied (and hence be unable to deliver the full 100 m³ of water when required) over the next 100 day period. The primary output of the model is simply the volume of water in the pond as a function of time.This volume, however, is uncertain (and must be simulated probabilistically) due to the variability of the inflow rate, and the random nature of the fires (i.e., they occur approximately once every 50 days).

To build this model, we will need create elements The basic building blocks with which a GoldSim model is constructed. Each element represents a feature, pararamer, process or event in the model. representing the pond, the inflow rate, the leakage rate, the occurrence of fires, and the sudden withdrawal of water when a fire occurs.We will also need to create an element the tracks the lowest volume of water observed in the pond over the time frame of the simulation.

To build this model, we are going to walk through all the steps necessary to build and run the model, including:

• Creating a new model and defining the simulation settings;
• Adding elements and defining their properties;
• Linking the elements together;
• Running the model; and
• Viewing and formatting the results.