?

Log in

No account? Create an account
   Journal    Friends    Archive    Profile    Memories
 

Systems Engineering Challange - RonO's Ramblings

May. 12th, 2008 01:18 pm Systems Engineering Challange

At and after this weekend's Karate tournament, I've been thinking off and on about if and how technology could be used to improve the scoring at a similar event. After a bit of thought, I narrowed it down to a problem in systems engineering or system design.

At a National Karate Chicago tournament -- the only ones I have experience with so far -- they set up 12 regular rings and 3 overflow rings. At each ring there are three or four judges scoring form, self defense, basics and similar events. Each judge awards points out of ten, but in reality one of 50 scores between 9.50 and 10.00 and usually from a more restricted range decided by the judges after counting the competitors and before starting the event. These are added together and the top three scores become the top three places, with the remaining competitors becoming "finalists." (There is some effort to make sure that one judge doesn't overwhelm the others, mostly by trying to minimize the number of different scores that can be issued. We also are instructed to not give out the same score twice so that there is a clear tiebreaker, at least with three judges).

The judges are joined by a pair of scorekeepers, one who reads the judges scores to the other scorekeeper and the audience, and the other who records them and adds them up.

The obvious first step for automation would be to replace the paper form with a spreadsheet on a laptop or tablet computer. However, the expense for each scoring position currently is a clipboard and a cheap calculator (both of which may be donated since one of the usual sponsors is an office supply chain). Therefore, to make it useful any solution would have to cost much less.

Therefore, it occurred to me that the real solution would be something a lot less expensive. However, the market for scoring machines for Karate tournaments -- and I believe that these systems would be portable to most North American Sport Karate Association (NASKA) tournaments -- would be limited. So supply and demand would make them unfordable. Which led me to the thought of an inexpensive device that could be loaded with a number of single purpose applications.

Pretending for a moment that I'm writing a project problem for a Systems Engineering course, I've come up with the following:

Our customer requires a system that will load and run a single, but customizable, application. This application may require quick entry of both text and numeric data, and will require a graphical or text output on a display with a resolution of at least 320 pixels, by 200 pixels each pixel supporting at least 8 bit color (256 colors). The launch application will be a custom spreadsheet like application that will require the entry of names and scores, the summing of scores and the reporting of the top scoring entries.

Other requirements:

  • Each system can have no more than a purchase price for hardware of $50.00 at full production
  • Each station should be able to operate on internal power for at least 8 hours without charging or replacing cells
  • Each system should be capable of operating independently, but will be sold in minimum lots of 5.
  • If needed a loading station, costing no more than $300.00 at full production, may be required at purchase
  • The systems should meet {mumble standard} durability specifications
  • The system should support an open, preferably standard, application language.

Some additional "nice to haves" would be: ability to network and communicate with a network printer or similar device, ability to load different applications.

Should any actual instructor wish to use this, they have my permission

I've heard that there are, or may be, some processors that more or less directly implement the Java Runtime Environment. If such a beast was available, it would be ideal for this application. If not, Java could still be used, but the core system would have to then start the JRE.

I think that 320x200 8bit LCD displays should be fairly inexpensive in production quantities, possibly with touch screen capabilities. Obviously a touch screen of sufficient size would allow there to be no other keys, just a virtual keyboard that appears when needed.

For loading applications, the best method might be simply to supply an enclosed area that can house a single USB flash drive which would contain the application in a .jar file in the root directory.

To make this work, there would be a fair amount of software would have to be put onto the device outside of the custom application, but it would mostly be made up of the device drivers and boot software. Any comments?

Leave a commentPrevious Entry Share Next Entry