(640b) Improving the Environmental and Economic Performance of Industrial Processes Using a Multi-Objective Optimization Framework

Sabio, N. - Presenter, Universitat Rovira i Virgili
Guillén-Gosálbez, G. - Presenter, University Rovira i Virgili
Jiménez, L. - Presenter, Universitat Rovira i Virgili
Vasudevan, V. - Presenter, Corporate Strategic Research, ExxonMobil Research and Engineering
Karuppiah, R. - Presenter, Corporate Strategic Research, ExxonMobil Research and Engineering
Sawaya, N. - Presenter, Corporate Strategic Research, ExxonMobil Research and Engineering
Farrell, J. T. - Presenter, Corporate Strategic Research, ExxonMobil Research and Engineering
Pozo, C. - Presenter, Universitat Rovira i Virgili

In this work, we have developed a decision support framework for the combined environmental and economic optimization of industrial processes. The goal is to determine the optimal operating conditions for an industrial process that simultaneously maximize an economic indicator and minimize an environmental metric, given a certain product demand, quality specification, and cost data. We have used a complex industrial process to demonstrate the application of the methodology.

We use life cycle assessment (LCA) principles (Guinée et al., 2002) to quantify the environmental performance of the industrial system under study. Life cycle assessment is a methodology that is used to evaluate the environmental performance of products, processes, or services, and is increasingly being incorporated into process / product design, R&D, and policy/regulations. A key benefit of LCA is that it accounts for inputs (mass and energy use) and outputs (emissions) over the entire life cycle, from raw material extraction to manufacture, use, and end-of-life, that is from cradle-to-grave. As such, it avoids shifting environmental burdens from one part of the supply chain to another, and facilitates a fair and holistic comparison of product / process options.

The optimization problem is mathematically formulated as a multi-objective nonconvex mixed-integer nonlinear program (MINLP) that incorporates economic and environmental criteria in an explicit manner. In this formulation, the binary variables are associated with discrete decisions such as choosing a certain type of utility, while the continuous variables correspond to operating conditions such as temperature, pressure, flow and composition of the process streams. The model constraints include standard, simplified empirical process correlations for representing the operation of different process units in the facility. Several cases that differ in final product demands, as well as the metrics to be optimized are run with the given model. For each case study, we generate Pareto optimal curves that show a set of alternative solutions, representing the trade-off between economic and environmental objectives.


Guinée, J. B.; Gorrée, M.; Heijungs, R.; Huppes, G.; Kleijn, R.; de Koning, A.; van Oers, L.; Sleeswijk, A. W.; S. Suh, S.; de Haes, H. A. U.; de Bruijn, H.; van Duin, R.; Huijbregts, M. A. J. Handbook on Life Cycle Assessment. Operational Guide to the ISO Standards; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2002.