EECS 509 / IOE 517 - Course Syllabus, Winter 1997

Course Topics:

  1. Introduction: transportation networks, traffic flow, travel demand, traffic assignment, route guidance, traffic signal control, Intelligent Transportation Systems.
    [1 class]
  2. Fundamentals of Traffic Flow
    1. Link flow theory: modeling of traffic flow on an individual link.
      Fundamentals of traffic flow: variables of interest, basic flow-speed-density relationship ("fundamental equation"), models of traffic flow (e.g., Greenshields, Greenberg, May).
      Introduction to microscopic car-following models: linear car-following models, asymptotic and local stability, steady-state behavior, nonlinear car-following models, steady-state behavior.
      Introduction to macroscopic fluid-flow models: continuity equation, recovering Greenberg's model, propagation of disturbances (density waves), shock waves.
    2. Traffic analysis at intersections.
      Case of signalized intersections: deterministic arrivals, stochastic (Poisson) arrivals, approximate analysis (based on M/D/1 queues) of Webster and Allsop and corresponding empirical formulas for delay.
      Case of unsignalized intersections: analysis based on gap acceptance and M/G/1 queues.
      Case of vehicle-actuated intersections.
      Platoons and platoon dispersion models.
    3. Impedance functions used as link/link+node performance functions in network optimization: Bureau of Public Roads function, Davidson's function.
    [10 classes]
  3. Traffic Signal Control
    1. Fixed-cycle signal timing plans at individual intersections: case of deterministic arrivals, case of stochastic (Poisson) arrivals, allocation of "free time" and equisaturation "rule", Webster's rule for cycle length.
    2. Fixed-cycle signal timing plans and progression method: case of one-way arterial, case of (symmetric) two-way arterial, left and right interferences, algorithm of Brooks, network optimization with fixed-cycle plans, TRANSYT.
    3. Demand-responsive signal control of individual intersections: method of rolling-horizon, optimization for one horizon, decision tree and tree pruning, introduction to shortest-path algorithms in decision networks (Dijkstra's algorithm and A*).
    4. Demand-responsive signal control of a network of intersections: general approach ("group-based" control), heuristic approaches (e.g., SCOOT, SCATS, ALLONS-I).
    [9 classes]
  4. Route Guidance and Traffic Assignment in Networks
    1. Introduction to route guidance and traffic assignment, user equilibrium and system optimality.
    2. Route guidance: static case, turn penalties, dynamic case, consistency condition and shortest-path algorithm.
    3. The static traffic assignment problem.
      Viewpoint of traffic assignment.
      Building the problem formulation: link-flow and path-flow formulations, multiple origin-destination pairs, cost model and link performance functions, criteria of optimality and Wardrop's principles.
      Special cases of the static traffic assignment problem.
      General static traffic assignment problem: necessary and sufficient conditions for user equilibrium and system optimality, marginal costs, transformation technique for solving for user equilibrium, Braess's paradox.
    [7 classes]
  5. Traffic Data Generation: introduction to origin-destination travel demand estimation.
    [if time permits]

Recall that the four books mentioned above are:

Back to course homepage


Last Modified: 1997/05/05