December:
Newsletter Home  I ProModel Home I Product News  I Santa Model  I Tips and Techniques

Tips and Techniques
Here is the PKG file that is used with this tip: multi unit resource.PKG. You must have a version of ProModel in order to run this file.

Multi-Unit Resource Variable Shift Assignment

Understand the problem
Very often, we need to determine the number of resources a system requires to achieve target levels of performance. In addition, we need to find out what their schedule should be.

Typical applications
ProModel: Deciding how many operators are needed to run a manufacturing cell and what their schedule should be.
ServiceModel: Deciding how many tellers are required to run a bank office and what their schedule should be.
MedModel: Deciding how many nurses are required to help patients in a hospital and what their schedule should be.

When to use this technique
You may want to use this technique to determine how many resource units (out of a maximum number of resource units) you need, and the shift to which each should be assigned.

ProModel allows you to test different shift and resource unit combinations by creating a multi-unit resource and defining possible shifts for such resource. For instance, if you define a ten-unit resource, and there are three possible shifts you would like to evaluate, then you would need to create thirty shift records, or three for each resource unit. Then, by creating a record-range-type macro for each resource unit, you can enable and disable shift records and thus define scenarios. You need to go to Simulation | Scenarios to be able to run the model. This method sounds somewhat tedious, doesn’t it?

This technique proposes an alternative method. This method is particularly useful if you need to manually test different scenarios.

When to avoid using this technique
This technique should not be used if you plan to use SimRunner to find an optimum solution. If that is the case, then it would be better to follow the method described above.

Statistics issues
Make sure you collect statistics “By Unit” instead of by “Summary” to avoid biasing the summary statistics with non-utilized resource units.