> I wrote: > > IIRC, there was some serious research on this, from UK, a couple of > > years back, which concluded that while early merging seems > > counter-intuitive to the driver it actually improves the overall > > throughput. And also that the theory assumed a certain percentage of > > assholes that seemed too low to me. > > > > I'll see what Google can turn up. > > 20 minutes of looking around didn't turn up the scientific reference, > unfortunately. However I found this readworthy piece which explains my > thoughts a lot better than I could have done it myself: > > http://blog.case.edu/singham/2007/06/08/highway_merging_and_th > e_theory_of_evolution > > Jostein
One thing about merging early rather than late that may affect the result is that when you merge as late as possible there is only one merging point. If people merge early there tends to be more than one, so it is less likely to impede traffic flow. However, road signs do instruct drivers to merge now - they artificially create a single merge point well before it is strictly necessary, so all the problems associated with merging are just pushed further back from the real bottleneck. This is where it's important to distinguish between how you merge, and when you merge. How you merge is most fairly and efficiently done in zip-like fashion by each driver giving way to one other driver at the merge point. When you merge, assuming you do it zippily, determines how much of the road is wasted - the earlier you merge the more of the road you waste. People who pass early mergers to gain an advantage are perceived by the early mergers to be cheating. But the 'cheating' is because they are not allowing anyone else to merge ahead of them. You could merge as early as you like but if you didn't let anyone merge ahead of you you would be cheating in just the same way. There was a piece about this on the BBC website last year. See item 7. http://news.bbc.co.uk/1/hi/magazine/7635486.stm This chap is some sort of expert on traffic behaviour, although I don't know whether or not he is a professional. http://www.google.co.uk/search?hl=en&q=Tom+Vanderbilt+traffic&meta= Last week I googled some more scientific papers from road research labs. Without having read more than an abstract of any of them it appears that the researchers are as divided about this as everyone else. Bob -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
