Does anyone know of a study that compares Simtraffic's modeling of a roundabout with empirical data or other analysis-based software such as rodel/sidra?

Thanks.

For U.S. roundabouts, I'm using the NCHRP 572 formulas now. In simulation I would only include the roundabout to have traffic move from one location to another, and not report the simulated results of the roundabout. For urban arterials I still prefer simtraffc, but the roundabout modeling needs more options, and improvement before it is as flexible as it needs to be, in my opinion.

I posted an almost identical question a few months ago. I have still not come accross any studies. If you do, please share. I believe right now, trafficware will tell you not to use the model for roundabout analysis. We use sidra and vissim, although I have coded in roundabouts like Eric said, just to move traffic in a network