Because transportation systems involve massive complex human activities, there exist substantial unpredictable uncertainties of the traffic demands. This paper aims at presenting an H ∞ control method for transportation network that can enhance the tolerance of the system due to these uncertainties. In particular, the store-andforward approach is applied to model the system into a linear form. Then, a detailed controllability analysis shows that the system is not completely controllable by taking the constraints on the green times into account. This makes difficult to apply directly the H ∞ method. To overcome this difficulty, this paper isolates the fully controllable part of the transportation system, and the problem of disturbance attenuation is then solved by means of a convex optimization with linear matrix inequality. Finally, the simulation of a large-scale hypothetical network is carried out to illustrate the results.Apparently, by introducing the nominal situation, the term ω has been defined as the differences between the real unknown traffic demands and the nominal ones. In other words, ω reflects the unknown 438 H. ZHOU ET AL.