Sunday, April 6, 2008

Grid Analysis

Hi again folks, I'm back again. I got coffee brewing up in the MIS lab so if your still awake and have nothing better to do, come join me. I have nothing better to do than blog. But enough of this shanidegans, lets get into this shall we. According to http://www.mindtools.com/pages/article/newTED_03.htm
Grid analysis can also be know as Decision Matrix, Pugh Matrix or MAUT (Multi-Attribute Utility Theory) if those ring a bell for anyone reading this. What it does is help you decide between several options. So keeping that in mind, lets set up a scenario and play a little game. Come with me to see how deep the rabbit hole really is. Wow I really need sleep. Say you want to go on vacation after you graduate UMSL (finally). You have saved up $1500 for a week stay someplace at you need help deciding where. You have narrowed it down to Las Vegas, Orlando, Hawaii, New York, Las Angeles, or a very nice stay in the St. Louis area (Ozarks included). Well you make up a grid and list the cities on the left and criteria you wish to consider up top, like price, weather, fun, places you want to see, etc. You give each city and criteria a score and weight and then total it up and the city with the highest score is your new vacation spot. I'm sure you all looked at the site and have seen the example grid. But thats about it. Of course each score is up to you and if your still unhappy with your decision, as the article suggests, giving a higher wieght to a criteria will sway your decision. Only this about this that I have a problem with is that if you don't have a definite score/wieght for each criteria, then you might spend more time deciding what each score/weight needs to go where.

1 comment:

Vicki said...

Such approaches only work when criteria are compensatory. That is when a low value on one dimension can be made up by a high value on another dimension, and where there are not any hard constraints. Once you add weighting, you increase the assumptions. Finally, this only works if you can assign a value that means something to your options.