On the simplicity side, simulation is as down-to-earth and natural, as children playing; they are in fact simulating a world of their own with their toys, objects and other children. I guess that’s why I love doing simulation studies, because I get to play in somebody else’s sand box, all the while getting to see the ingenuity from some of our most creative minds. Yes, it’s also a form of continuous education as I believe that at least 50% of the benefits you get out of a simulation study are the things you learn along the way, which makes sense. If you think about how complex systems are designed today, the work is not performed by a single entity or group, but rather by many sub groups working together to each build their own part of the puzzle. You might have mechanical, electrical, constructional, quality, reliability and many other kinds of engineers all working together, each specializing in their own field and each performing different tasks. The simulation specialist must gather and collect great deals of information from many groups just to build a credible model and subsequently gains a broad understanding of the system.
Once the initial model has been built it must be verified and validated, which require that all interested parties come back together to ensure that the model does mimic the system under study and the output from the model is representative of what we would expect from the real system. This process, in its self, often causes a build upon ideas mentality where the system designers discover better ways to approach their design and therefore make continuous improvement suggestions for it. In this way the model is used to further explore their system. Our experience has shown us that most studies may go through several iterations of improvement and fine tuning until the stake holders have a high degree of confidence in the tool and consequently their design.
The model is now ready for execution; this is where the purpose of the study comes into play. Various model runs are made and analyzed by changing one parameter at a time to determine its affects on the system. Sensitivity analysis is one tool used for determining the point of diminishing returns for a particular parameter. For instance if the model is of a flexible manufacturing system we might want to know how many technicians do we need; if they do all the maintenance, tool changes and part handling activities. We would make a series of model runs, each of which would increase the number of technicians by a small amount then plot the results.
Furthermore, many customers today are asking for more out of their models, especially as computing power has increased and simulation software has improved. Typically this requires the modeler to ask questions of the designers and engineers for things which are often not thought of until later in the process.
We are finding that clients today are asking for studies to gain deep insight into a process for the purpose of:
· Validating a proposed process such as a new equipment design
· Improving existing processes
· Studying a process for the purpose of discovering key characteristics or attributes which affect particular parts of a process
· Developing best practices
· Training by un-cluttering complexity to show key aspects
I’ve often heard clients ask for a simulation of some process, what they are really trying to say I need some process or design studied for some purpose. Sometimes the study will be of a proposed or theoretical physical system and sometimes it is of an existing system which appears to be performing at a level less than optimal. In either case, the conducting of the study goes through similar phases.
Best Regards,
John
