The fundamental thinking underpinning process intensification (PI) involves an understanding of process phenomena at various levels and their interrelationships without the spatial constrains of conventional unit operation models. This paper presents a framework in which the process models are based on the physicochemical phenomena arranged into abstract (i.e. equipment-independent) functional, structural and behavioural modules. This modularisation framework is used in a methodology to encourage the generation of intensive design strategies. Topological representations are first generated from qualitative knowledge and then mapped into mathematical models. Causal graphs are introduced to allow the designer to identify the relationships between variables relevant to the process. The models are implemented in object-oriented software to quantify the net effect of selected design variables or to quantify equipment requirements. This approach is proposed to increase the flexibility, customisation and reusability of models in order to encourage the implementation and assessment of novel intensified and multifunctional process options. © 2007 Elsevier B.V. All rights reserved.