Microscopic and Macroscopic Approaches to Delay Estimation with Oversaturated Conditions

Chapter 16 of the 2000 Highway Capacity Manual (HCM) contains a widely recognized and well-accepted procedure for calculating per-vehicle control delay at signalized intersections. Appendix F of that chapter discusses the relationship between various components of control delay using cumulative arrival and departure curves. The delay obtained from these curves has certain limitations that need to be recognized if proper conclusions are to be drawn. To illustrate these limitations, a comparison is made between the delay obtained from cumulative curves and the delay obtained through a detailed analysis of vehicle trajectories. A comparison of the control delay obtained from trajectory analysis and that obtained from cumulative arrival–departure curves shows that the cumulative curves omit certain valid portions of the control delay, while including other portions of time that are not delay at all. Taking a macroscopic approach to oversaturated delay estimation, the HCM procedures for dealing with oversaturated conditions are described and their shortcomings are explored using an example of a multiperiod analysis of an oversaturated approach to a signal. It is demonstrated that the current HCM equations can overestimate multiperiod delay and that the assumption of a capacity that is constant from cycle to cycle can underestimate delay at high values of the volume-to-capacity ratio.