The usual way that these type of things are modelled is through thought experiment games.
So take the scenario of Vehicle to Grid energy swapping, what would normally happen is that you create different scenarios and then ask people to put a 'price' on different aspects of them. Then look at the data and see what comes out.
One scenario may be:
Allow up to 40% discharge, limiting the range of the vehicle to 60 miles.
Then you ask one group of people how much they want to be paid for that, and ask another group how much they would have to pay to use an alternative if they had to (say they need the car to do more than 60 miles).
With large enough sampling they two figures should match.
Generally what happens is that these sorts of surveys are biased from the beginning.
So another survey would offer 3 different payments to allow the DNO to discharge your car, but offer no alternative method of transport.
Another could offer you 'free' public transport in return for discharging.
Sampling bias is the biggest problem, this happens a lot when they do surveys to 'value nature'. They do a lot of them down here via the nearby university. These surveys often happen at my nearby park. The trouble is that the respondents are a self selected group by the very act of being where the questions are being asked about.
So if we did a survey on here about V2G we would get very different answers and some people would have to be excluded from the start, no vehicle owners, people with no alternatives (that is Damon and myself out, Damon don't want a car and there are no alternatives for me to get home from work).
This is a major problem when trying to establish if an idea would actually work. Especially true when there are cultural and geographical differences.
Edited by joiner, 06 March 2016 - 09:01 AM.