Multicriteria optimization of gluconic acid production using net flow

Abstract

The biochemical process industry is often confronted with the challenge of making decisions in an atmosphere of multiple and conflicting objectives. Recent innovations in the field of operations research and systems science have yielded rigorous multicriteria optimization techniques that can be successfully applied to the field of biochemical engineering. These techniques incorporate the expert's experience into the optimization routine and provide valuable information about the zone of possible solutions. This paper presents a multicriteria optimization strategy that generates a Pareto domain, given a set of conflicting objective criteria, and determines the optimal operating region for the production of gluconic acid using the net flow method (NFM). The objective criteria include maximizing the productivity and concentration of gluconic acid, while minimizing the residual substrate. Three optimization strategies are considered. The first two strategies identify the optimal operating region for the process inputs. The results yielded an acceptable compromise between productivity, gluconic acid concentration and residual substrate concentration. Fixing the process inputs representing the batch time, initial substrate concentration and initial biomass equal to their optimal values, the remaining simulations were used to study the sensitivity of the optimum operating region to changes in the oxygen mass transfer coefficient, K(L) a, by utilizing a multi-level K(L) a strategy. The results show that controlling K(L) a during the reaction reduced the production of biomass, which in turn resulted in increased productivity and concentration of gluconic acid above that of a fixed K(L) a.

Comparison of in‐situ recovery methods of gas stripping, pervaporation, and vacuum separation by multi‐objective optimization for producing biobutanol via fermentation process

“…The Pareto domain for the continuous fermentation integrated in turn with each of the three separation methods was obtained by determining 1000 Pareto-optimal solutions using the Dual Population Evolutionary Algorithm (DPEA). [23] These 1000 solutions were then ranked using the net flow method based on the relative weights and threshold values of each criterion presented in Table 3. The plots of the three or four objective functions for the ranked Pareto domains for gas stripping, pervaporation, and vacuum fermentation are presented in Figures 4, 6, and 7, respectively.…”

“…In this paper, the Dual Population Evolutionary Algorithm (DPEA) was used to generate the Pareto domain for each scenario analyzed in this paper. [23] The DPEA is a simple and fast evolutionary algorithm that has been shown to be very robust and equivalent to other more popular evolutionary algorithms such as NSGA-II. [27] The general approach to obtain the Pareto domain is briefly described as follows:…”

Continuous gluconic acid production by Aureobasidium pullulans with and without biomass retention

“…Park et al (2000) cloned and expressed for example glucose oxidase from A. niger in Saccharomyces cerevisiae using a yeast shuttle vector. Specialized reactors (Basseguy et al 2004;Godjevargova et al 2004) and genetic programming, as well as multicriteria optimization strategies (Cheema et al 2002;Halsall-Whitney et al 2003) have been applied for the production of gluconic acid as well. The fact that the present superior results were reached using a wild strain as it has been isolated from the nature and without the involvement of any classical or genetic engineering mutagenesis approaches clearly emphasize the nature's great latent and still unknown potential for further future biotechnological achievements.…”

Multiobjective flux balancing using the NISE method for metabolic network analysis

“…Several attempts to solve multiobjective problems have been made. A multiobjective optimization strategy was used to determine the optimal operating region for production of gluconic acid by net flow method (Halsall-Whitney et al, 2003) and to perform simultaneous optimization of several responses in the case of ethanol production by Saccharomyces cerevisiae (Vera et al, 2003). The bilevel optimization framework, OptKnock, was introduced to propose reaction eliminations from the E. coli network for maximizing the production of simple compound such as succinic acid, lactic acid, and 1,3-propanediol , and for the enhancement of amino acid overproduction .…”