View Single Post
  #25   Spotlight this post!  
Unread 17-03-2006, 12:02
Daniel_LaFleur's Avatar
Daniel_LaFleur Daniel_LaFleur is offline
Mad Scientist
AKA: Me
FRC #2040 (DERT)
Team Role: Engineer
 
Join Date: Apr 2005
Rookie Year: 2003
Location: Peoria, IL
Posts: 1,964
Daniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond reputeDaniel_LaFleur has a reputation beyond repute
Send a message via MSN to Daniel_LaFleur
Re: American Inventor

Quote:
Originally Posted by Mercury Rising
I have put some thought into it and realized why this wouldn't work and/or sell. Not only would a tiny fan do absolutely nothing to the 140 degrees that a car can heat up to but even if it could the items in the car will still be hot. First off you would need a fan the size of the window and second off something that big could be smashed in and used to break into cars. Also with this product there is the chance of a failure, which in this case if there is the user will die. Having a product like that must be perfect, and will be too expensive to be plausible. Not only that people will be convinced if it is good enough for a dog it is good enough for a baby, now a dog dying because of a failure is one thing but a baby is a terrible terrible lawsuit.
Its called a DFMEA (or design failure mode and effect analysis). DFMEA rates the effect, how often it occurs, and how easy it is to detect (before the failure happens, this is usually if it can be detected in the factory). These numbers are multiplied together to get a DFMEA rating.

In the above case effect would be a 10 (death of user), so for this to be a viable product the occurence would have to be very low (under 1 part per 10 million, or ~2) and the failure detection would have to be almost flawless (also ~2). Neither of these are likely. Most automobile companies work with DFMEA numbers under 80.
Reply With Quote