Modelling Queuing of Vehicles at Signalized Intersection

  • Dhaval Parmar
  • Ninad Gore
  • Dipak Rathva
  • Sanjay DaveEmail author
  • Manish Jain
Conference paper
Part of the Lecture Notes in Civil Engineering book series (LNCE, volume 45)


The present study emphasizes on the determination of the queue length at the signalized intersection by manually collected data and modelling queue length for a busy urban signalized intersection under heterogeneous traffic conditions of Vadodara city, Gujarat, India. Major approaches and one of the minor approaches have an identical width. Flow share of 65% was observed on major approaches. Flow composition on major approaches was characterized by the presence of heavy vehicles; whereas, the same was absent on minor approaches. Moreover, red time of both major and minor approaches was articulated according to flow share. Each lane has a minimum two-lane width but the queue length was not respective to lane discipline. Queue length was measured manually for two hours in the evening peak for three days by graduating medians at 1 m interval. In addition, queue length and its composition were recorded simultaneously until the last second of red time for each lane. Observations revealed that the average queue length varied from 71.47 to 110.64 m on the major approach while it varied from 46 to 55 m on the minor approach, respectively. Further, queue length for one of the major and one of the minor approaches exceeded the cross traffic opening and hence hindered the free movement of cross traffic. Queue composition was dominated by motorized two-wheelers followed by cars and motorized three-wheeler (auto-rickshaws). It was also noted that though queue length was similar, the total PCU value per cycle varied on one of the major approaches. This may be attributed to the presence of heavy vehicles in queue composition resulting into the large gap between vehicles, reducing its local density, and thus reflecting driver behaviour. It was also observed that queue length was dependent upon its composition and associated red time. Multi-linear regression analysis was used to model queue length with respect to its composition, associated red time and width of the road. Further, the queue model for three legs of identical width was statistically validated against queue model of one remaining leg to examine the effect of road width using F-test and was found insignificant. Predicted queue length values were also checked with observed queue length values using t-statistics and t-test analysis, which shows there is no significant difference between two data sets. Mean absolute percentage error (MAPE) was estimated around 14%, indicating fair acceptance of developed queue model.


Signalized intersection Queue length Queue model 


  1. 1.
    Anusha Lelitha SP, Vanajakshi D, Sharma A (2013) A simple method for estimation of queue length. Digital Commons University of Nebraska—Lincoln Civil Engineering Faculty PublicationsGoogle Scholar
  2. 2.
    Dey PP, Nandal S, Kalyan R (2013) Queue discharge characteristics at signalized intersections under mixed traffic conditions. Eur Transp 8(55). ISSN 1825-3997Google Scholar
  3. 3.
    Jagannathan K, Modiano E, Zheng L (2011) On the role of queue length information in network control. J IEEE Trans Inf Theory 57(9):5884–5896MathSciNetCrossRefGoogle Scholar
  4. 4.
    Evans L, Rothery RW (1981) Influence of vehicle size and performance on intersection saturation flow. In: Proceedings from 8th international symposium on transportation and traffic theory. University of Toronto Press, Toronto, Ontario, pp 193–222Google Scholar
  5. 5.
    Liu HX, Wu X, Ma W, Hu H (2009) Real-time queue length estimation for congested signalized intersections. Transp Res Part C 17:412–427CrossRefGoogle Scholar
  6. 6.
    Chang J, Talas M, Muthuswamy S (2012) A simple methodology to estimate queue lengths at signalized intersections using detector data. Resubmitted to Transportation Research Board for possible publication in the Transportation Research RecordGoogle Scholar
  7. 7.
    Lee S, Wong SC, Li YC (2015) Real-time estimation of lane-based queue lengths at isolated signalized junctions. Transp Res Part C Emerg Technol 56CrossRefGoogle Scholar
  8. 8.
    Wu A, Yang X (2013) Real-time queue length estimation of signalized intersections based on RFID data. In: Proceeding of 13th COTA international conference of transportation professionals (CICTP 2013)Google Scholar
  9. 9.
    Ban XJ, Hao P, Sun Z (2011) Real time queue length estimation for signalized intersections using travel times from mobile sensors. Transp Res Part C Emerg Technol 19(6)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Dhaval Parmar
    • 1
  • Ninad Gore
    • 1
  • Dipak Rathva
    • 1
  • Sanjay Dave
    • 1
    Email author
  • Manish Jain
    • 1
  1. 1.The Maharaja Sayajirao University of BarodaVadodaraIndia

Personalised recommendations