dc.description.abstract | Increasingly, many roundabouts in Kampala city are becoming problematic due to capacity challenges resulting from rapid traffic growth. Signalisation is known to improve capacity, shorten queue length, and reduce delay at roundabouts but this option has not been used in Uganda. In this study, two computer models namely signalized and un-signalized were explored for analysis and comparison of their performance made with Kubiri roundabout as a case study. The aim of the study was to establish which model option suits best in terms of improved capacity, reduced delays and queue lengths. Field traffic volume studies for the entire roundabout’s approaches were performed; travel time and delay studies for the subject approach (Gayaza approach) were also undertaken. Geometric data as well as data from the field studies described earlier were used to calibrate a model for the existing situation. Capacity, delay, queue length and degree of saturation were observed for the signalised options and existing un-signalised. The results indicated that the full signalisation model produced the better parameter results as compared to the un-signalised model. However, there is an increase in the queue length and delay as well as a decrease in capacity on Gayaza approach attributed to alteration in the balance of traffic flow due to the signalisation on all the approaches as well as the circulatory carriageway. Generally, there is a slight improvement in the flow of traffic after signalisation however that the level of service is still low i.e. LOS F and degree of saturation is still high (greater than 1.00). This could be attributed to the fact that the total pcus from all the approaches exceed 10000 pcus/hr which necessitates grade separation. | en_US |