Modelling of Bicycle Operations at Signalized Intersections

Abstract

The bicycle is an effective mode of transportation is highly essential for developing country to overcome the several issues like traffic congestion, shortage of petroleum fuel, accident rate, air pollution level etc. Due to lack of bicycle facilities, use of bicycles has been reducing with time. There are no special facilities for bicycle users like bicycle lane and wide outside lane etc. for which the bicyclists face so many problems in the highly mixed traffic conditions. Signalized intersections at which traffic signals are installed to control the traffic movements and regulated by the traffic police or by signal signs. These are the critical section of a road section having many conflicts and merging points. Due to the mixing of slow and high moving vehicular performances, delay for bicyclists increases for which it causes discomfort for bicyclists. The accurate estimation of delay is difficult due to the random flow of traffic, but the prediction of delay is very important. For better bicycle infrastructure, the existing facilities should be assessed in a systematic manner for which this may cause the increase of bicycle level of service criteria and decrease of delay measure effectively.
In this study, the data has been collected from the 60 signalized intersection approaches of the mid-sized cities in India named as Bhubaneswar, Nagpur, Lucknow, Ranchi, Tirupati, Anantapur, Kurnool. The overall scores signifying the satisfaction levels perceived by the road users has been obtained in this study. Pearson correlation analysis has been carried out for identifying the significant variables and four variables are identified as significant for BLOS and two variables are identified to be important for the delay. The four significant variables of BLOS are approach width (AW), pedestrian volume (PV), cross peak hour volume (CPHV), delay (D). The two significant variables for delay are capacity (c), percentage of heavy vehicles (HV). Using these significant variables, the non-linear regression analysis has been developed and tested using the statistical parameters such as correlation of determination (R2), error measuring parameters (AAE, MAE, RMSE). Thus, the significant variables have utmost important roles to play in the development of bicycle-friendly signalized intersections.