Optimal Chemical Product Design via Fuzzy Optimisation based Inverse Design Techniques

Lik Yin Ng, Nishanth G. Chemmangattuvalappil*, Denny K. S. Ng

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapter

10 Citations (Scopus)


While designing optimal products using molecular design technique for a specific application, product superiority is the only factor considered during design. It is noted that the effectiveness of this approach is highly dependent on the model accuracy which represents product robustness. Thus, to design an optimal product efficiently, the effect of property prediction uncertainty has to be addressed. This paper presents a systematic methodology for the design of optimum molecules used in chemical processes by considering both product superiority and robustness. Fuzzy optimisation approach is incorporated into inverse design techniques for this purpose. Inverse product design techniques utilize molecular design techniques to predict the molecular structure of the product from the target properties. Optimal solution is selected based on how much the solution satisfied the criteria of product superiority and robustness. The effectiveness of the developed methodology is shown by a case study on fungicide design, where the designed fungicide possess superior attributes than the conventionally used fungicide.

Original languageEnglish
Title of host publication24th European Symposium on Computer Aided Process Engineering
Number of pages6
ISBN (Print)9780444634566, 9780444634344
Publication statusPublished - 2014

Publication series

NameComputer Aided Chemical Engineering
ISSN (Print)1570-7946


  • Fuzzy optimisation
  • Inverse design techniques
  • Product design
  • Property prediction uncertainties

ASJC Scopus subject areas

  • General Chemical Engineering
  • Computer Science Applications


Dive into the research topics of 'Optimal Chemical Product Design via Fuzzy Optimisation based Inverse Design Techniques'. Together they form a unique fingerprint.

Cite this