Foffa Bikes

No, you would be making something which your model predicts will be comfortable. Unless you're just making a direct copy of an existing product, all your measurement based on existing models may or may not be useful - you might have just accidentally measured only those parameters which reliably predict for comfort on existing product, but have missed some vital aspect which is only shown up by the failure of your new product in testing.

A prediction based on objective measurement. Testing would have to be an integral part of the entire process. And measurements, data gathered, adjusted accordingly. Hypothetical of course but it is possible to treat something as subjective as the creation of 'comfort' in an objective way. Major manufacturers do it every day.