Re: New Car for Jenn
Posted: Sat Nov 18, 2017 7:10 pm
Ok I threw this together.
I just started with some basic requirements and put my own weights on them. 3 of the 5 requirements require normalization which I take care of, given a reasonable best/worst range (the other worksheet). Don't type in the gray fields.
For example, power/weight. I figured a high water mark for this purchase is about 400 hp to 3850 lbs (which translates to CTS-V1 performance). And the low water mark around be around 100 hp and 3200 lbs (Ranger with a passenger). The actual value of the vehicle under evaluation scores from 0-10 depending on where it falls in this established range.
I did the same thing for EPA combined fuel economy (2008+ standards) and interior volume.
You want the values to be somewhat realistic bounds or the resulting score won't be sensitive enough to these values (i.e. if we looked from lawn mower to top fuel dragster in power/weight, all of these vehicles would rate almost the same and it would make no difference). So take a look at those.
Then you end up with a raw score and a weighted score.
The idea is, you can tweak around with the weights to properly reflect preferences. Try to put some dynamic range in there unless she really doesn't care about one more than another. And you can add requirements as necessary. If they are quantitative they will need to be normalized from 0-10.
What I did not factor in was pricing or availability. In other words, if you guys fill this out, and you find a winner, it is up to you to find an example that exists and is affordable. I assume you wouldn't enter a vehicle in the list if it was a totally impossible proposition.
Besides, price and availability are variables. The things captured in this matrix are, hopefully, constants. The good news is, if preferences shift and one thing becomes more or less important, that can be changed and all vehicle scores would be recalculated, which is cool.
Here is a OneDrive link to the file for group editing (use this, not the attachment): https://1drv.ms/x/s!An5lt1BKvFKdqKg-nYJxkNN2ILrgRA
I just started with some basic requirements and put my own weights on them. 3 of the 5 requirements require normalization which I take care of, given a reasonable best/worst range (the other worksheet). Don't type in the gray fields.
For example, power/weight. I figured a high water mark for this purchase is about 400 hp to 3850 lbs (which translates to CTS-V1 performance). And the low water mark around be around 100 hp and 3200 lbs (Ranger with a passenger). The actual value of the vehicle under evaluation scores from 0-10 depending on where it falls in this established range.
I did the same thing for EPA combined fuel economy (2008+ standards) and interior volume.
You want the values to be somewhat realistic bounds or the resulting score won't be sensitive enough to these values (i.e. if we looked from lawn mower to top fuel dragster in power/weight, all of these vehicles would rate almost the same and it would make no difference). So take a look at those.
Then you end up with a raw score and a weighted score.
The idea is, you can tweak around with the weights to properly reflect preferences. Try to put some dynamic range in there unless she really doesn't care about one more than another. And you can add requirements as necessary. If they are quantitative they will need to be normalized from 0-10.
What I did not factor in was pricing or availability. In other words, if you guys fill this out, and you find a winner, it is up to you to find an example that exists and is affordable. I assume you wouldn't enter a vehicle in the list if it was a totally impossible proposition.
Besides, price and availability are variables. The things captured in this matrix are, hopefully, constants. The good news is, if preferences shift and one thing becomes more or less important, that can be changed and all vehicle scores would be recalculated, which is cool.
Here is a OneDrive link to the file for group editing (use this, not the attachment): https://1drv.ms/x/s!An5lt1BKvFKdqKg-nYJxkNN2ILrgRA