An integrated approach to prioritise parameters for multi-objective optimisation: A case study of biomass network. (20th November 2020)

• Record Type: Journal Article
• Authors: Yeo, Shi Zhuan; How, Bing Shen; Ngan, Sue Lin; Ng, Wendy P.Q.; Leong, Wei Dong; Lim, Chun Hsion; Lam, Hon Loong
• Is Part Of: Journal of cleaner production. Volume 274(2020)
• Publication Date: 2020-11-20
• Subjects: Multi-criteria decision-making -- Multi-objective optimisation -- Principal component analysis -- Analytical network process -- Integrated approach; Factory and trade waste -- Management -- Periodicals; Manufactures -- Environmental aspects -- Periodicals; Déchets industriels -- Gestion -- Périodiques; Usines -- Aspect de l'environnement -- Périodiques; 628.5
• DOI: 10.1016/j.jclepro.2020.123053
• Languages: English
• ISSNs: 0959-6526

Abstract

Abstract: Decision-making can be a complicated task when contradicting factors have to be considered and the complication is extended when qualitative factor is to be included. Mathematical approaches have been developed to generate a range of solutions for optimal decision to be made. In this work, Principal Component Analysis (PCA) and Analytic Network Process (ANP) are employed to develop sets of optimal decisional parameters which consider both the quantitative and qualitative inputs. PCA is a multivariate statistical tool that has been widely used for dimension reduction and to identify the importance of parameter based on its variation. However, this prioritisation method overlooks the customised needs of the decision-makers. ANP is an approach that is capable of deriving the priority weightages of a set of parameters based on the importance, influence power and dependency relationship of the elements. In order to achieve an optimal solution, max-min aggregation fuzzy technique is applied onto the solution sets to tighten the range of possible solutions and to determine an optimal variable which fits the multi-objective considered. A theoretical case study is presented to demonstrate the application of the proposed two-stage methodology. The demonstration proposes an effective approach to generate a narrowed range of optimal solutions that considers both quantitative and qualitative elements for decision-making. Highlights: Multi-objective optimisation provides a range of feasible solutions for decision-making. The range of optimal solutions can be tightened to reduce the options to ease decision-making. Multi-objective model can be optimised by considering both quantitative and qualitative elements. Min-max aggregation approach is used to integrate the objectives to generate a tightened range of optimal solution. The proposed optimisation method aggregates the outcomes of PCA and ANP efficiently tightens the range of feasible solutions.