This paper presents a systematic approach to efficiently reconstruct the infinite dimensional field in distributed process systems from a limited, and usually reduced, number of sensors. To that purpose, two basic tools are employed: on the one hand, a reduced order representation of the system which, based on proper orthogonal decomposition (POD) expansions, captures the most relevant dynamic features of the solution. On the other hand, the selection of the most appropriate type (and number) of measurements by the solution of a max-min optimization problem. These ideas will be illustrated on the problem of field reconstruction for unstable tubular reactors. (C) 2003 Elsevier Ltd. All rights reserved.