How easy is it?
Using SimKit is easy. Soon you will be running models like the one below and learning important details about your system that could make the difference between success and failure.
How do I start?
All the systems that we are interested in can be broken down into activities. In the main these will be processes (that is, tasks) and queues. But there will also be other more specialised activities such as combiners and dividers, for example. In SimKit terms all these activities, whatever their nature, are called components. You simply connect them together to build your system – just as you would for real. The above is a model of a simple flexible manufacturing system or cell to illustrate how straightforward models can be. Unfinished blanks are fed onto a conveyor belt. In general terms these are called entities and represented by SimKit as single characters. At some point they are diverted off the line into a machine that converts it to finished parts. At another point the finished parts are diverted off the line into a packing process. Unfinished parts circulate around until the machine is available. This is a simple system with one single machining process. But it could easily be expanded to include more. Incidentally, this system is not very efficient. Can you tell why?
There must be more to it than that?
Not much!
The only other thing you have to do is supply simple parametric information about all the components you use, such as process times. Then you just run the model, watch and gather the data. Incidentally, you have lots of gadgets available to display information as the model runs, like process activity and buffer/queue content monitors. And you can display animations so that even your boss can understand it. If you really want to be wicked, you can use sound files for special effects as well!
At the end of the simulation run, SimKit provides you with a Result Analyser so that you can display your data in a variety of chart forms. But if you want it, the numbers are there as well.
The first step, then, is building a model. How do I do that?
Models are built on screen in your virtual workspace simply by placing symbols that represent system activities or components and then interconnecting them. If you have previously used a charting application you will already be familiar with this technique albeit in a different context. As you might expect, the symbols can be moved around until you get a tidy arrangement.
The symbols are chosen by name from a pop-up menu which shows all the components in the current library. Actually, a library is simply a list of all the components that you want to use chosen from all that are available. Naturally, you can make up your own library if you want. And adding new components is simple too. But all that is explained in SimKit’s extensive document set.
Interconnecting the components of your virtual system is made easy by SimKit turning them into terminal blocks, the screw heads being the ‘hot spots’. This ‘wiring’ is done point to point but SimKit then automatically routes it orthogonally to give a neat and easily readable schematic. The model that you build is saved to a Configuration file.
The next step is supplying parametric data. How do I do that?
As well as configuration data, all systems have associated parameters. For example, the average time that you estimate a task will take to complete or the maximum length of a queue that you are going to tolerate. Each component of your virtual system has an associated utility which will prompt you for all necessary information as you proceed. You reach this by simply double-clicking on a component’s symbol. In practice you will probably want to generate several sets of this parametric data as you explore your virtual system’s behaviour and tolerance to different circumstances. In consequence, SimKit allows you to arrange all your parametric data in sets of Parameter files.
Incidentally, you can supply the name of a file here too. This could be a manufacturer’s data sheet for a particular item of equipment that the component represents. That means, you can have available at your finger tips all the information you need.
Do I get to run the model now?
Absolutely!
So far you have provided all the Information. This is now where SimKit takes over and generates behavioural data on how your system is expected to perform, simply by clicking your mouse on a button.
Naturally you can pause the simulation and re-start again at any time you like so that you can examine points of interest as it progresses. A SimKit simulation run can generate lots of information so great emphasis is placed on providing you with what you want in an appropriate way.
As you start the simulation you will see that the passive component symbols that make up your model change to active icons capable of displaying information in lots of different ways. This is a model of an in-line production system.
Displays include numbers – naturally – as well as histograms and in-context animations. Animations are particularly helpful in conveying the function and purpose of your system to others who are less familiar with system design than you are.
Other utilities are also provided to allow you to see in more detail what is going on during the simulation. These include task/process activity and queue/buffer content monitors. There are also facilities to halt the simulation when certain conditions occur.
If you pause the simulation or it halts for any reason all accumulated behavioural information is written to a Report File. This file is the primary information source for further analysis. Numbers are fine but graphic displays are a good way to appreciate complex data. Accordingly, SimKit provides you with a tool to examine and present generated data in chart form. The displays are not confined to the current run. In fact, you can compare charts obtained from previous runs so that you can better understand the behaviour of your system under different circumstances.
Of course, this is only a flavour of what you can do with SimKit. Fortunately, SimKit is shareware, so you can try it out for free to see if it will cover your needs. The only limitation is that some of its more advanced features may not be available until you register your copy. So either click on the button on the bottom of this page or on the link in the contents list to reach the down-load section. There is information there to tell you what to do and what to expect.
Do I need a powerful system to run SimKit?
Definitely not.
SimKit runs under all current versions of Microsoft Windows, i.e., Win95/98, NT4 and 2000, but we do recommend a Pentium 200 and 32MB of RAM for optimum performance and a Hard Disk space of about 40MB for the system software, help files and other text based documentation and application notes. Remember, because this is essentially a graphical application, you will benefit from a good monitor which should be as large as possible.
In addition, please note that the full suite of technical documentation for SimKit is available online at its support site SimKit.co.uk.
(c) ROSIM Systems.