Lesson 10 - Unit 9 Summary

In this Unit, we explored the complex dynamics that can be generated by delays.

Delays are common in all real-world systems. Delays represent the reality that in many cases, if you do something now, it's impact will not be immediate, but will be delayed.

This Unit illustrated how delays can generate complex dynamic behavior in models. 

The key points that we covered were as follows:

• GoldSim provides two different elements to represent delays: Material Delays and Information Delays.
• Material delays represent the delay in the movement or processing of material (e.g., water, items, people) through a system or process. For example, when you mail a letter, there is a delay before it is delivered.  When you start flowing water into a pipeline, there is a delay before it exits the pipeline. When you put a meal into an oven, there is a delay before it is cooked.
• Information delays represent the delay in the collection and/or transmission of information. For example, when you take a measurement of some variable that is required to make a decision, perhaps it takes some time (hours, days) before that information is properly collected and transmitted.  As a result, any actions that are taken in response to that information are not actually based on the present state of the system, but a past state of the system.  
• In addition to simply representing delays in the physical movement (flow) or in some cases the change of state of a system, Material Delays are also critical for handling recirculating (recursive) logic in a model.
• Feedback control systems that use delayed information as the process variable can perform poorly.  If indeed your system has such reporting delays, the Information Delay element allows you to properly represent this.
• Information Delays can be used to simulate forecasts. By defining the Bias of a Proportional Controller using a forecast, you can greatly improve the performance of the Controller.

Now that we have covered the fundamentals of modeling material flows and have explored how to represent some of the complex dynamics that often occur in such systems, we are ready to tackle some of the more advanced concepts in GoldSim. We will start doing that in the next Unit by discussing how to create hierarchical models, a necessity for creating a model of even moderate complexity.