Abstract:Bio-inspired algorithms are a suitable alternative for solving multi-objective optimization problems. Among different proposals, a widely used approach is based on the Pareto front. In this document, a proposal is made for the analysis of the optimal front for multi-objective optimization problems using clustering techniques. With this approach, an alternative is sought for further use and improvement of multi-objective optimization algorithms considering solutions and clusters found. To carry out the clusteri… Show more
“…Similar work using variant clusters to guide swarm optimisation has also been published in the fields of computation and automated learning. 31,32 Figure 4.Design search method 2: Fully automated global search followed by local search using RNSGA, k-means and silhouette score. Each step is supported with a Radviz type visualisation of the output.…”
Section: Methodsmentioning
confidence: 99%
“…Similar work using variant clusters to guide swarm optimisation has also been published in the fields of computation and automated learning. 31,32 The final phase of this methodology identifies a final set of solutions from the many iterations. For each set of result generated by R-NSGA-II, a subset of weakly-dominated and non-dominated solutions is identified using pareto sorting and the method described above is reapplied to this subset (Figure 4).…”
Section: Workflows For Design Exploration Using Optimisation and Mach...mentioning
Embedding circular economy (CE) principles in early design requires iterative evaluation across multiple lifecycle phases, with trade-offs between objectives complicating the identification of best solutions. This paper puts forward methods to automatically discover diverse, yet well-performing solution types within complex multi-objective CE design optimisation models. Working with a parametric model derived from a furniture design for CE case study, a comparison is made between weighted-sum single objective optimisation and multi-objective optimisation augmented with clustered solution types targeted by the reference point-based NSGA-II optimisation algorithm. Efficiency of optimisation, quality of results and distinctiveness of solution types presented by each method is compared in an effort to understand which will best assist designers to manage complexity in CE design. The generalisability of the presented methods to larger scale CE design problems is discussed and future areas of work on computational design for CE are extrapolated from the presented results.
“…Similar work using variant clusters to guide swarm optimisation has also been published in the fields of computation and automated learning. 31,32 Figure 4.Design search method 2: Fully automated global search followed by local search using RNSGA, k-means and silhouette score. Each step is supported with a Radviz type visualisation of the output.…”
Section: Methodsmentioning
confidence: 99%
“…Similar work using variant clusters to guide swarm optimisation has also been published in the fields of computation and automated learning. 31,32 The final phase of this methodology identifies a final set of solutions from the many iterations. For each set of result generated by R-NSGA-II, a subset of weakly-dominated and non-dominated solutions is identified using pareto sorting and the method described above is reapplied to this subset (Figure 4).…”
Section: Workflows For Design Exploration Using Optimisation and Mach...mentioning
Embedding circular economy (CE) principles in early design requires iterative evaluation across multiple lifecycle phases, with trade-offs between objectives complicating the identification of best solutions. This paper puts forward methods to automatically discover diverse, yet well-performing solution types within complex multi-objective CE design optimisation models. Working with a parametric model derived from a furniture design for CE case study, a comparison is made between weighted-sum single objective optimisation and multi-objective optimisation augmented with clustered solution types targeted by the reference point-based NSGA-II optimisation algorithm. Efficiency of optimisation, quality of results and distinctiveness of solution types presented by each method is compared in an effort to understand which will best assist designers to manage complexity in CE design. The generalisability of the presented methods to larger scale CE design problems is discussed and future areas of work on computational design for CE are extrapolated from the presented results.
“…Therefore, a solution point converging to one of the available objectives moves away from the other. Multi-objective optimization issues in this situation can only be solved [18]. It is not always possible to guarantee that one suggested strategy outperforms the others.…”
In this study, a new approach based on the Smith chart is proposed for multi-objective particle swarm optimization. With this approach, the position of each particle is updated according to the Smith chart’s impedance and admittance circles to determine the local best particle. The proposed method is applied to three test functions with different behaviors, namely concave, convex, and non-convex, and performance parameters are examined. In light of the obtained results, it has been observed that the proposed strategy offers successful convergence performance for multi-objective optimization applications and meets performance expectations with its obtained well-distributed solution set.
“…Alternative approaches, such as clustering, have been shown to improve the convergence and diversity of Pareto optimal solutions. For example, the selforganizedspeciation-based algorithm has been proposed to solve multimodal multiobjective problems by avoiding overlap between species to obtain more evenly distributed Pareto optimal solutions [24], while the k-means or fuzzy cmeans algorithm has been used to assist the multiobjective vortex particle swarm optimization (MOVPSO) algorithm in easily identifying the individual center of the swarm [25]. However, these approaches rely on predetermined values for parameters, such as the radius or number of clusters, which can make the model infexible to changing distributions of solutions.…”
In multiobjective particle swarm optimization (MOPSO), the global-best particle is randomly selected for each population particle from a nondominated solution set. However, this Roulette wheel-based global particle selection is ineffective for convergence and diversity when the problem has numerous decision variables or a large number of global-best candidates. Thus, this study proposes the cluster-based MOPSO (CMOPSO). In CMOPSO, the similarities between particles are considered when selecting the global-best particle. The cluster for each particle is determined based on the Euclidean distance in the decision or objective space. The proposed approach is demonstrated by applying an operating condition optimization problem to the hydrogen production process. The target process is a representative chemical plant with a large search space and strong nonlinearity. Furthermore, the performance of CMOPSO is assessed by comparing it with that of MOPSO. The results indicate that CMOPSO considered in the decision space exhibits superior performance in terms of convergence and diversity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
