This paper presents the implementation of a large-scale systems analytic method in a model permitting the assessment of air quality policies in urban regions. This method is based on the use of a convex optimization technique called the analytic center cutting plane method. One realizes the coupling of a photochemical model (TAPOM-Lite), used to simulate ozone creation and dispersion in the region under study, and of a technoeconomic model (MARKAL-Lite) that represents the technology and energy uses by different economic sectors in the same region. Although the models correspond to different time and space scales, one realizes the coupling through a series of approximating convex reduced order optimization problems with constraints that are implicitly defined by the photochemical and the technoeconomic model, respectively. The implementation of the method in a case study corresponding to the region of Geneva, Switzerland is described. (C) 2003 Elsevier Ltd. All rights reserved.