## Typical questions that might come up in oral exams on traffic dynamics and simulation

• Summarize the contents of traffic flow dynamics [as displayed on the ten sets of slides] in a few words
• What is traffic flow dynamics good for? give some use cases
• Determine macroscopic quantities such as traffic flow, density, average speed, propagation velocities of traffic waves from a sketch or plot of trajectories that will be handed out during the exam
• What is the difference between trajectory data as extracted from fixed cameras, FC data, and stationary detector data (SDD)? Give the strenghts and weaknesses of each data category with respect to some use cases
• A stationary detector detects six passages of a vehicle within a sampling period of one minute and records an average speed of 36 km/h. Give the estimated flow in vehiclesa/hour and density in veh/km. Why is the density estimate probably biased?
• What is a flow-density scatter plot and a fundamental diagram (FD)? Draw a typical fundamental diagram and flow-density data containing both free and congested traffic
• (advanced) Explain the Simpson effect when analyzing a day of traffic flow data
• (advanced) [Given a sheet of paper with some data points of vehicle trajectories (FCD) and stationary detector data (SDD) in the x-t plane]. How data inconsistencies may arise when both SDD and FCD are available? How to resolve this issue?
• Discuss the difference between microscopic and macroscopic traffic flow models
• Discuss the differences between time-continuous models, iterated maps, and cellular automata. Is it possible to formulate a macroscopic model by a cellular automaton?
• How to derive the hydrodynamic relation flow=speed times density?
• Write down the simplest continuity equation (homogeneous road) on a sheet of paper. Now integrate both sides of this equation over space from x1 to x2. What do you get?
• Discuss the Eulerian and Lagrangian point of view
• Write the simplest LWR on a sheet of paper and discuss the terms and variables
• Given a LWR model with a certain fundamental diagram on a sheet of paper and two points p1 and p2 on it.
• How to determine local propagation velocities at a density corresponding to p1?
• Given is a transition from p1 to p2 while traffic is in state p1 upstream and in state p2 downstream of the transition region, respectively: Under which condition a shockwave will form? If it forms, what will be its velocity?
• Give examples of traffic-related bottlenecks (the more, the better)
• A traffic jam with a sharp transition free to congested forms upstream of a bottleneck. How to determine if the jam area increases or shrinks based on a counting detector at the bottleneck and a detector in the free-flow region further upstream? (Notice: may also come as a numerical example)
• Give to mechanisms leading to a resolution of an existing traffic jam (from the demand and supply side)
• A partial road block caused by an accident and causing a congestion is removed. Give the spatiotemporal dynamics of the congested region
• How to macroscopically model traffic lights?
• (advanced) explain the idea behind the supply-demand method when numerically integrating LWR models
• Give the difference between micro and macroscopic traffic flow models
• Discuss in words the plausibility criteria for the acceleration function f(.) of a time-continuous car-following model
• (advanced) [Given a simple time-continuous or iterated-map car-following model such as dv/dt=(v0-v)/tau - lambda1/s + lammbda2*(vl-v) with speed v, leading speed vl, gap s]
• Discuss the meaning and typical values of the model parameters v0, tau, lambda1, lambda2
• Determine the homogeneous steady state/the fundamental diagram
• Check the plausibility criteria
• Discuss the relation between a car-following model and the core of an adaptive-cruise control
• Which human characteristics are not considered by the Intelligent-Driver Model? (primarily the finite reaction time but also many more)
• Discuss the rough idea behind the derivation of the Gipps model and the IDM (no calculations!)
Back to the lecture page

Martin Treiber