View source for Opasnet:New thoughts and discussions

=R90(?) Research: Composite traffic in the European Union=

==Objective of the study==

The objective of this research is to evaluate the impacts of a putative composite traffic system that covers the whole European Union (EU27). Composite traffic means a system where a centralised system collects information about all trips and aggregates the trips into vehicles based on demand. The main focus is on integration of different traffic modes at all trip lengths from 1 km to 5000 km. The comparison is to the current situation of car, air, train, bus, and ship traffic. Here we assume that the trip volumes and share of each traffic mode remain the same, but the number of vehicles needed and kilometres driven will change. 

==Study design==

This is a modelling study that is based on a fixed number of trips on an origin-destination matrix for one working day. European trips will be modelled based on easily available overall statistics. 

There should be a way to model trips on various levels: 
*>1000 km European (business, vacation) 1M trips per day
*200-1000 km national (business, holidays) 5 M trips per day
*20-200 km regional (commuting, business) 40 M trips per day
*<20 km city (commuting, daily activities) 1200 M trips per day

There should be a way to model these different levels at some degree of detail. Should it just be separate models for each model, and a separate model for each city size? This implies some kind of general model. 

Location of people in a grid:
* The size of each cell is a random variable ~lognorm(1, gsd)
* The sum of all cells is scaled to the total number of people.
* The number of trips between two grid cells is proportional to the product of their sizes, i.e. the probability of a destination is proportional to the size of the cell.
*Different city structures can be tested by changing gsd: a low value results in a homogenous city.

If we use the A4-fractale approach, it can be used to any scale. The problem is that it should be realistic enough so that kilometres driven can be compared with the real statistics. Can we model the current situation with a A4-approach? If yes, then the comparison must be based on (käyttöaste) use efficiency (how large fraction of  the available seats are in use).