Discrete Event Model

This is the first of two videos on a discrete event model for a printed circuit board shop. Here we are just treating the plant as a conveyor belt where the expenses are accounted by looking at daily amounts of money going out each day. You can find the companion article for this video here. Discrete event model remains an underutilized tool in understanding the potential within an operation and discovering the levers which you can control. The primary emphasis here is cleaning up the model so that it is more visual, it is better organized, and makes intuitive sense. In this way people can interact with it better and ask more probing questions which is key to learning. Discrete even modeling is a key advanced feature we offer clients and can be part of the deliverables in a DAM Blast.



Here we show a very simple DOE to determine the capacity of the plant. In order for the model to work and match Little’s Law, we have to take into account subtleties such as the warm up period. When we do that then we get results that are accurate. The output to production problems are typically highly nonlinear, so when doing a DOE on a simulation you have to collect lots of data, which just costs processor time. How you analyze the data is also important and in this case I took the Data from my ExtendSim model and fed it into JMP. I then use two different methods of fitting this data that nearly matches the theoretical result. This video and the first video on this discrete event model, emphasizes the need for validating any model. This is a critical part of any modeling building process.


Resource Constraints

Cost accounting does not handle resource constraints! Because of this it is very easy to get the wrong answer and start making decisions that are highly damaging to your business. The most powerful and efficient way of identifying those resource constraints is using linear programing. In this example, we see how the resource constraint changes based on product mix. We also discover that what defines the most profitable job for the plant isn’t the one with the highest margin, but the one that can generate the highest dollars per unit time through the constraint.


Unproductive Infighting

The impossible problem plagues any business.  The impossible problem often isn’t impossible for business leadership, such as get the next generation of product out the door and commercialized, but seems impossible for your people to do, because they just aren’t getting along.  What causes all of this unproductive infighting where, as an example, manufacturing and research and development, argue passionately against each other?  If your company has the money, do you have to hire teams of psychologists?  No.  Your people aren’t crazy, it’s the environment your people are in that makes them act crazy.  Here we introduce some of the ideas we use to develop much more collaborative team based constructs.


1 Mil Lines & Spaces & Beyond

This video brings in three industry heavy weights to discuss the seemingly impossible problem of 1 mil line and spaces and beyond. This is a companion to the feature PCB Magazine article, “1 Mil Lines & Spaces & Beyond: Reclaiming Technological Leadership.” This video is an example of when you feel your boss has asked you to do the impossible, you can always find the help you need if you are willing to ask the seemingly “stupid” questions. You can only ask for help if you don’t have an ego DAM.

1 Mil lines & Spaces & Beyond

Discrete Analyzing Measurex Data



No comments yet.

Leave a Reply