The aim of this work is to understand how urban traffic behavior, especially in cases
of congestion, can be improved by an accurate choice of traffic coefficients. For this, we define three
cost functionals that measure average velocity, average travelling time and total flux of cars. The
global optimal control problem for a complex network is difficult to solve both from analytical and
numerical points of view. Thus, we focus on a simple junction with one incoming road and two
outgoing roads (junctions of 1×2 type), obtaining exact solutions to a simple optimization problem.
Then, we use such results at each node of the network. The traffic evolution of some networks is
then studied via simulations. In particular, it is shown that an appropriate choice of the traffic
distributions can be useful in order to improve network conditions.