A hierarchical agent-based approach to simulate a dynamic decision-making process of evacuees using reinforcement learning

Hassanpour, Sajjad; Rassafi, Amir Abbas; González, Vicente A.; Liu, Jiamou

doi:10.1016/j.jocm.2021.100288

Cited by 12 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, they noticed that students who follow the premeditated instructions leave the classroom faster and are more successful in ensuring their own safety during evacuation procedure. Another pedestrian micro-simulation study focusing on pedestrian evacuation process belongs to Hassanpour et al (2021). Similarly, Wang and Jia (2021) presented a tsunami evacuation risk assessment.…”

Section: Netlogomentioning

confidence: 99%

Etmen-tabanlı modellemede belirme ve karmaşıklık: Güncel araştırmaların incelenmesi

Cenani

2021

Journal of Computational Design

View full text Add to dashboard Cite

Agent-based systems are an important application area of artificial intelligence and are used in decision support systems. Rather than being a problem-solving tool, agent-based system is a tool for developing and testing alternative solutions according to various scenarios. In this context, agent-based modeling is a very effective method to support decision makers in emergency situations to evaluate different risk scenarios and then make decisions quickly and effectively. Moreover, agent-based modeling is a very useful method to support decision makers in situations of high complexity and uncertainty. This paper introduces current studies performed with several agent-based modeling tools and software environments such as NetLogo, AnyLogic, MATSim and Repast. Apart from these, various agent-based modeling tools exist, but these four tools have been chosen because they are still receiving software updates and being widely used in the most current studies. The aims of this study are to review these four agent-based modeling tools, present state-of-the-art research conducted with these tools and provide a reference of agent-based modeling tools for researchers who are developing decision support systems in architectural, urban and transportation design research fields. In this paper, after giving a brief definition of an agent-based system and explaining the importance of concepts such as emergence and complexity in the field of agent-based modeling, it is explained who uses the agentbased models for what purpose, when, where, why and how to use agent-based modeling through selected examples from state-of-the-art studies carried out in different research fields. Furthermore, what current studies and agent-based modeling tools teach us and how future studies can benefit from agent-based models are briefly discussed.

show abstract

Section: Netlogomentioning

confidence: 99%

Etmen-tabanlı modellemede belirme ve karmaşıklık: Güncel araştırmaların incelenmesi

Cenani

2021

Journal of Computational Design

View full text Add to dashboard Cite

show abstract

“…Second, RL can help in modelling adaptive agent behaviour(Gazzola et al 2016;Li et al 2019;Ramchandani et al 2017). Third, RL can be used to model human decision-making more realistically (Al-Khayarin & Halabi 2021;Hassanpour et al 2021;Li et al 2020;Pang et al 2018). For instance, Al-Khayarin & Halabi (2021) apply RL to emulate the behaviour of people in the real-world in the context social distance measures.…”

mentioning

confidence: 99%

Using Machine Learning for Agent Specifications in Agent-Based Models and Simulations: A Critical Review and Guidelines

Dehkordi

Lechner

Ghorbani

et al. 2023

JASSS

View full text Add to dashboard Cite

Agent-based modelling and simulation (ABMS), whether simple toy models or complex data-driven ones, is regularly applied in various domains to study the system-level patterns arising from individual behaviour and interactions. However, ABMS still faces diverse challenges such as modelling more representative agents or improving computational efficiency. Research shows that machine learning (ML) techniques, when used in ABMS can address such challenges. Yet, the ABMS literature is still marginally leveraging the benefits of ML. One reason is the vastness of the ML domain, which makes it difficult to choose the appropriate ML technique to overcome a specific modelling challenge. This paper aims to bring ML more within reach of the ABMS community. We first conduct a structured literature review to investigate how the ABMS process uses ML techniques. We focus specifically on articles where ML is applied for the structural specifications of models such as agent decision-making and behaviour, rather than just for analysing output data. Given that modelling challenges are mainly linked to the purpose a model aims to serve (e.g., behavioural accuracy is required for predictive models), we frame our analysis within different modelling purposes. Our results show that Reinforcement Learning algorithms may increase the accuracy of behavioural modelling. Moreover, Decision Trees, and Bayesian Networks are common techniques for data pre-processing of agent behaviour. Based on the literature review results, we propose guidelines for purposefully integrating ML in ABMS. We conclude that ML techniques are specifically fit for currently underrepresented modelling purposes of social learning and illustration; they can be used in a transparent and interpretable manner.

show abstract

A hybrid hierarchical agent-based simulation approach for buildings indoor layout evaluation based on the post-earthquake evacuation

Hassanpour

González

Liu

et al. 2022

Advanced Engineering Informatics

View full text Add to dashboard Cite

A hierarchical agent-based approach to simulate a dynamic decision-making process of evacuees using reinforcement learning

Cited by 12 publications

References 21 publications

Etmen-tabanlı modellemede belirme ve karmaşıklık: Güncel araştırmaların incelenmesi

Etmen-tabanlı modellemede belirme ve karmaşıklık: Güncel araştırmaların incelenmesi

Using Machine Learning for Agent Specifications in Agent-Based Models and Simulations: A Critical Review and Guidelines

A hybrid hierarchical agent-based simulation approach for buildings indoor layout evaluation based on the post-earthquake evacuation

Contact Info

Product

Resources

About