A surrogate approach for the global optimization of signal settings and traffic assignment problem

We extend a `surrogate problem´ approach that is developed for a class of stochastic discrete optimization problems so as to tackle the global signal settings and traffic assignment combined problem. We compare a stochastic method based on the surrogate approach, called Surrogate Method (SM), with a Projected Gradient Algorithm (PGA), which uses the Armijo rule for the step size estimation routine. Numerical experiments conducted on a test network show that the surrogate method converges to a really small area and it finds much more efficient solutions.