A modeling framework is proposed for complex chemical process networks. The state is represented by energy, volume and mass inventories. The dynamic behavior of the process system is constrained so that all trajectories satisfy the first and second laws of thermodynamics. The concavity of the entropy function is used to define a storage function for passivity design. The proposed storage function is closely related to the Gibbs tangent plane condition. A multi-component version of Tellegen's theorem from circuit theory is used to develop sufficient conditions for stability of process networks. The sufficient conditions can be interpreted as dissipation conditions for production and flow. These results can be used to design decentralized inventory and flow control systems for process networks of arbitrary complexity. Flow and inventory control are introduced for various process units, including a stirred tank reactor and a flash unit. We develop a plant wide control system for a recycle problem with a reactor and a distillation column. (C) 2007 Elsevier Ltd. All rights reserved.