(397a) Optimal Design of Bioprocesses with Economic and Environmental Concerns Via a Combined Simulation-Optimization Approach
In the last years, bioprocesses have become increasingly important, given their potential to produce high-value products in human health and care applications. The recent boost in competitiveness for customers and new products experienced in this sector has led to an increasing interest in bioprocess modeling and simulation techniques.
The literatute that deals with the optimization of bioprocesses, including all their individual steps, is quite scarce. In fact, nowadays the design of these processes is typically assisted by empirical and/or intuitive methods such as rules of thumb or simple heuristics that are likely to lead to sub-optimal solutions (Koulouris et al., 2000; Petrides et al., 2006). Furthermore, aside from systematizing the search for optimal solutions, there is the issue of how to account for multiple conflictive criteria at the design step. Particularly, with the recent trend of developing more sustainable processes, it has become increasingly important to include environmental concerns in the decision-making procedure. Despite the effort made so far in this area, the current situation is that almost all the existing environmentally conscious process design tools have a narrow scope as they tend to focus at the plant level, thus neglecting the impact caused in other stages of the life cycle of the process. This limitation can lead to solution in which the environmental problem is just transferred to other echelons of the supply chain.
This work aims to fill in this research gap by introducing a new method for the design of biotechnological processes that allows for the simultaneous consideration of economic and environmental concerns at the design step. The methodology presented relies on combining simulation packages, optimization tools and life cycle analysis (LCA) principles within a unified and holistic framework.
The design task is mathematically formulated as a multi-objective mixed-integer dynamic optimization (MIDO) problem, which is solved by a decomposition method that iterates between primal and master sub-problems. The primal level of our algorithm entails the solution of a dynamic nonlinear programming sub-problem in which the integer decisions, mainly the type and number of equipments in parallel, are fixed. This primal problem is solved by integrating the batch process simulator SuperPro Designer with an external gradient-based optimization package. The master level is a customized MILP that is constructed by linearizing the objective function and constraints of the dynamic problem in its optimal solution with fixed binaries. This master MILP, which decides on the values of the integer variables, includes integer and logic cuts that exclude solutions explored so far as well as solutions that are likely to lead to worse objective function values. A heuristic-based procedure is used to derive the latter set of cuts based on the detection of bottle-necks in the flowsheet. The environmental impact is measured via the contribution to global warming following the CML 2001 methodology. The environmental analysis accounts for the damage caused by the extraction of raw materials, energy generation and direct emissions associated with the main process.
The capabilities of the proposed methodology are demonstrated through its application to the production of the amino acid L-lysine. Results show that it is possible to reduce the environmental impact by compromising the profit of the process. Particularly, our technique allows to identify non-intuitive operating conditions in the fed-batch reactor that can help in reducing the associated life cycle impact of the plant.
References. 1.Koulouris, A., Calandranis, J. and Petrides, D. P. (2000). Throughput Analysis and Debottlenecking of Integrated Batch Chemical Processes. Computers and Chemical Engineering. Vol. 24: 1387-1394. 2.Petrides D.P., Papavasileiou V., Kolouris A. and Siletti, C. (2006) Optimize manufacturing of pharmaceutical products with process simulation and production. Chemical Engineering Research and Design Volume 85, Issue 7, Pages 1086-1097