Many materials of engineering interest have highly heterogeneous microstruc
tures. To a first approximation, the response of multi-phase materials to e
xternal stimuli such as mechanical loading depends on global parameters suc
h as average particle size or phase volume fraction. Most classical models
of materials behaviour are based on such an assumption. It is clear however
that an accurate description must include parameters that characterize the
distribution of phases. Moreover, some processes that we wish to model are
inherently stochastic in nature. This adds considerable complexity. First,
the quantitative description of microstructure containing higher order mom
ents is fraught with difficulties - both analytical and experimental. Secon
d, the inclusion of clustering into analytical models is prone to assumptio
ns and approximations. In this paper we will restrict ourselves to phenomen
a for which a continuum approach is adequate. For these, self-consistent ap
proaches are especially valuable. The two examples that we discuss in some
depth are related to (i) damage in porous, brittle films such as thermal ba
rrier coatings and (ii) the simultaneous effects of damage and particle clu
stering on the elasto-plastic response of metal matrix composites. (C) 2001
Elsevier Science Ltd. All rights reserved.