A novel deceleration traffic flow model with oscillatory congested states

Abstract A novel deceleration traffic flow model is established based on the oscillatory congested states and the slow-to-start rule. The novel model considers human overreaction and mechanical restrictions as limited deceleration capacity, effectively avoiding the unrealistic deceleration behavior found in most existing traffic flow models. In order to consider that the acceleration of a stationary vehicle is slower than that of a moving vehicle due to reasons such as driver inattention, the slow-to-start rule is introduced. In actual traffic, the driver will take different deceleration measures according to local traffic conditions, divided into ordinary and emergency deceleration. The deceleration setting in the deceleration model with only ordinary deceleration is modified. Computer simulations show that the novel model can achieve smooth, comfortable acceleration and deceleration behavior. Introducing the slow-to-start rule can realize the first-order transition from free flow to synchronized flow. The oscillatory congested states enable a first-order transition from synchronized flow to wide moving jam. Under periodic boundary conditions, the novel model can reproduce three traffic flow phases (free flow, synchronized flow, and wide moving jam) and two first-order transitions between three phases. In addition, the novel model can reproduce empirical results such as linear synchronized flow and headway distribution of free flow below 1 s. Under open boundary conditions, different congested patterns caused by on-ramps are analyzed. Compared with the classic deceleration model, this model can better reproduce the phenomenon and characteristics of actual traffic flow and provide more accurate decision support for daily traffic management of expressways.