<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>This work presents a novel approach to the design of a decision-making system for the cluster-based optimization of an evacuation process using a Parallel bi-objective Real-Coded Genetic Algorithm (P-RCGA). The algorithm is based on the dynamic interaction of distributed processes with individual characteristics that exchange the best potential decisions among themselves through a global population. Such an approach allows the HyperVolume performance metric (HV metric) as reflected in the quality of the subset of the Pareto optimal solutions to be improved. The results of P-RCGA were compared with other well-known multi-objective genetic algorithms (e.g., -MOEA, NSGA-II, SPEA2). Moreover, P-RCGA was aggregated with the developed simulation of the behavior of human agent-rescuers in emergency through the objective functions to optimize the main parameters of the evacuation process.</p>
</abstract>

This work presents a novel approach to the design of a decision-making system for the cluster-based optimization of an evacuation process using a Parallel bi-objective Real-Coded Genetic Algorithm (P-RCGA). The algorithm is based on the dynamic interaction of distributed processes with individual characteristics that exchange the best potential decisions among themselves through a global population. Such an approach allows the HyperVolume performance metric (HV metric) as reflected in the quality of the subset of the Pareto optimal solutions to be improved. The results of P-RCGA were compared with other well-known multi-objective genetic algorithms (e.g., -MOEA, NSGA-II, SPEA2). Moreover, P-RCGA was aggregated with the developed simulation of the behavior of human agent-rescuers in emergency through the objective functions to optimize the main parameters of the evacuation process.

Cluster-Based Optimization of an Evacuation Process Using a Parallel Bi-Objective Real-Coded Genetic Algorithm

National Research University Higher School of Economics

Central Economics and Mathematics Institute of Russian Academy of Sciences

Cybernetics and Information Technologies

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

This work presents a novel approach to the design of a decision-making system for the cluster-based optimization of an evacuation process using a Parallel...

Cluster-Based Optimization of an Evacuation Process Using a Parallel Bi-Objective Real-Coded Genetic Algorithm

Online veröffentlicht: 13. Sept. 2020

Seitenbereich: 45 - 63

Eingereicht: 14. Mai 2020

Akzeptiert: 30. Juni 2020

DOI: https://doi.org/10.2478/cait-2020-0027

Schlüsselwörter
Cluster-based multi-objective optimization, human crowd behavior, real-coded genetic algorithms, multi-agent systems, fuzzy clustering

© 2020 Andranik S. Akopov et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Cluster-Based Optimization of an Evacuation Process Using a Parallel Bi-Objective Real-Coded Genetic Algorithm

Online veröffentlicht: 13. Sept. 2020

Seitenbereich: 45 - 63

Eingereicht: 14. Mai 2020

Akzeptiert: 30. Juni 2020

DOI: https://doi.org/10.2478/cait-2020-0027

SchlüsselwörterCluster-based multi-objective optimization, human crowd behavior, real-coded genetic algorithms, multi-agent systems, fuzzy clustering

© 2020 Andranik S. Akopov et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Schlüsselwörter
Cluster-based multi-objective optimization, human crowd behavior, real-coded genetic algorithms, multi-agent systems, fuzzy clustering