adjustments (Emigholz, 1996). By contrast, the operators of modern industrial plants face different challenges due to handling, analyzing, and interpreting large amounts of distributed information simultaneously.Although some advanced control techniques (e.g., dynamic matrix control, model algorithmic control) had their origins in the industry from some enlightened development groups (e.g., Shell, IDCOM), the formalization and systemization of these techniques was conceived, conducted, and finalized by academia through a rather long incubation period (see Morari & Lee [1999] for a detailed review). Eventually, starting from the 1990s, the availability of robust software environments and fast hardware, both reasonably priced, allowed the seminal model-based techniques (still too theoretical and simplified) to be transposed into stable and complex algorithms, such as model predictive control (MPC) and real-time optimization (RTO;De Souza et al., 2010).MPC and RTO enabled the industries to implement models and solutions into their systems by optimizing the process parameters to maximize the production according to minimum energy usage, and/or minimum raw materials consumption, and/or minimum environmental impact. The growth and implementation of these optimization techniques can be judged by a prediction made by the ARC advisory group, that the RTO market will reach more than US$1.5 billion in 2015 (Abel, 2011).