Design and analysis of roundabout operation was an issue for both traffic engineering practitioners and researchers until the aaSIDRA analytical based software evolved. Though it has the best performance among analytical models, it is not free from limitations in comparison with microscopic simulation models (i.e. PARAMICS). On the other hand microscopic simulation models tend to ignore capacity, the most widely used concept in traffic engineering practice. The reason behind is that, micro simulation models are based on different modelling paradigm (car following, queuing, acceleration etc). So it is often advised by researchers to calibrate microscopic simulation models using analytical models output data in the absence of real data. But to reflect the actual environment (unique geometry, public transport modelling) or adjacent effects, microscopic simulation model calibration can be also used to calibrate analytical models, which is one of the objectives of this study. Video observation has been conducted to collect approximate origin destination pattern and queue length. Model calibration parameters were systematically changed to assess their impact on the results with respect to field observations. The conclusion of this study suggests about the comparability of some of the performances and parameters for both models. At the recommendation stage, some issues have been raised in improving the comparability through better calibrations. (a) For the covering record of the conference, please refer to ITRD abstract no. E218380.
Abstract