Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application

ABSTRACT


INTRODUCTION 1.Research Background
One crucial aspect of managing structures like houses or buildings is the establishment of safety protocols to mitigate disaster risks.Disasters are a series of events that can happen at any time and can cause harm to humans, both material and non-material [1].The importance of these security measures lies in protecting life and property with a holistic approach that includes prevention strategies and emergency response measures.Efforts to identify and overcome potential disasters are fundamental to designing effective security systems for these buildings.Artificial Intelligence (AI) technology can minimise the fire risk in buildings.The integration of AI technology enables early detection, rapid response, and even realistic evacuation simulations to increase the efficiency of security systems and protect against disasters.AI is

Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application (Irfan Yusuf Muttaqien
essential for modern disaster security system planning.It can quickly process and analyse vast amounts of data, understand and predict human behaviour, adapt to real-time changing conditions, and continuously improve evacuation strategies.By leveraging AI, disaster response systems can become more efficient, effective, and capable of saving more lives during emergencies.The working process of AI involves collecting, merging, and sharing big data.Extensive data on the internet includes research, analysis, computing, and data storage required for a particular project.AI can create tools that significantly contribute to the longevity of architectural buildings.AI can increase efficiency in all building design and construction stages because it has high analytical and computational capabilities.AI can produce solutions to bring innovation to the world of architecture.Artificial intelligence directs architectural developments toward more sophisticated and practical solutions [2].Fields included in AI include agent-based modelling, expert systems, natural language processing, speech recognition, robotics, and artificial neural networks.
In an era where technology is increasingly crucial in everyday life, the use of AI Agent Base in disaster simulation reflects the integration of advanced technology to ensure public safety and security.Through the continued development of this technology, the hope is that future buildings will be safer and have greater resilience to potential disaster threats.A concrete example of implementing AI Agent-Based is the AnyLogic program, which represents how artificial intelligence-based solutions can improve safety and sustainability in the disaster sector.
AnyLogic is modelling and simulation software developed by XJ Technologies.This tool adopts the latest complex system design methodology and uses the latest system design methods.Its main advantage lies in introducing the UML language into the model simulation domain, making it the only software that supports a wide range of modelling methods with a high level of development [3].AnyLogic's ability to integrate factors into disaster evacuation simulations is a powerful tool for building designers and security experts.AnyLogic simulation provides an understanding of human behaviour in emergencies so that it is easy to overcome potential bottlenecks and the effectiveness of evacuation routes.AnyLogic is vital in developing realistic disaster evacuation simulations, significantly enhancing community safety efforts, and reducing disaster risks in various buildings.The software uses Agent-Based Artificial Intelligence (AI) to evaluate and improve the efficiency of emergency evacuation routes by modelling disaster scenarios, analysing human behaviour, and identifying bottlenecks.This research will help to improve safety, provide informed design decisions, and offer better insights into human behaviour during emergencies, ultimately contributing to more prepared structures and advanced disaster preparedness and response planning.

Problem Statement and Research Objectives
The problem statement in this research is "How can we review the use of Agent-Based AI to calculate the efficiency of emergency access to evacuation routes?"This research aims to review a program based on Agent-Based Artificial Intelligence called AnyLogic to calculate the efficiency of emergency evacuation routes.

RESEARCH METHODS
This research employs the journal review method as its primary approach, systematically collecting, reviewing, and compiling data from various relevant journal sources to present comprehensive information.By detailing, analysing, and drawing conclusions from the collected data, this method provides a strong foundation for an in-depth understanding of the researched phenomenon.It supports research sustainability through analytical and interpretive approaches.The research aims to explore the theoretical foundations and applications of Agent-Based Artificial Intelligence (AI), mainly focusing on AnyLogic's role in modelling emergency evacuation scenarios.
Key factors to be reviewed include AnyLogic's capabilities, previous case studies, and its effectiveness in disaster simulations.The research will assess human behaviour during emergencies, how AnyLogic models these behaviours, and identify potential bottlenecks in evacuation routes.Additionally, the study will synthesise findings to analyse the practical implications of AnyLogic in improving evacuation efficiency, supporting informed design decisions, and enhancing disaster preparedness and response planning.The research steps involve seeking theoretical bases, gathering literature on AnyLogic and disaster simulations, extracting essential insights, discussing findings, and drawing conclusions.This comprehensive review aims to provide a thorough understanding and meaningful conclusions about AnyLogic's role in emergency evacuation simulations.

Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application (Irfan Yusuf Muttaqien
This research contributes to the field by focusing on utilising AnyLogic, an Agent-Based Artificial Intelligence program, to assess the efficiency of emergency access evacuation routes.By employing a journal review method, the study delves into the capabilities of AnyLogic, analyses previous case studies, and evaluates its effectiveness in disaster simulations, offering valuable insights into optimising evacuation strategies.The novelty of this research lies in its detailed examination of how AnyLogic models human behaviour during emergencies, identifies potential bottlenecks in evacuation routes and aims to enhance evacuation efficiency, thereby providing a comprehensive understanding of the role of AI technology in improving safety measures in buildings during critical situations.

Theoretical Basis and Steps of AnyLogic Use 2.1.1 Disasters
As stipulated in Article 1 paragraph (1) of Law of the Republic of Indonesia Number 24 of 2007 concerning Disaster Management (UU PB), what is meant by disaster is an event or series of events that threatens and disrupts the life and livelihood of the community caused, by natural factors, and non-natural factors, as well as human factors, resulting in human casualties, environmental damage, property loss, and psychological impacts.Furthermore, Article 1 paragraph (2) of the PB Law explains that what is meant by natural disaster is a disaster caused by an event or series of events caused by nature, including earthquakes, tsunamis, volcanic eruptions, floods, droughts, hurricanes, and landslides [4].

Artificial Intelligence
Artificial Intelligence (AI) has the potential to become an inseparable part of people's daily lives.These changes impact personal experiences, changing how companies make decisions and interact with external stakeholders.AI technology can be integrated into various aspects of our routine.AI provides insightful solutions that simplify tasks and improve overall efficiency.The growing presence of AI in our lives represents a broader paradigm shift in society.Therefore, the increasing role of AI can change the routine and nature of decision-making processes.
Artificial intelligence (AI) is a branch of computer science that deals with developing computer systems and technologies capable of performing tasks that usually require human intelligence.AI aims to create programs or machines to learn, adapt, and perform tasks such as pattern recognition, natural language processing, decision-making, and problem-solving.AI can be used in various fields, including robotics, natural language processing, image recognition, and data analysis.The main goal of AI is to create systems that can think automatically and make intelligent decisions without human intervention [5].
Artificial intelligence (AI) is shaping a diverse landscape, from systems capable of interacting with humans to autonomous cars that enable driverless travel and even applications in the medical world to support clinical decision-making.It has significantly impacted various sectors, including health, automotive, and finance.AI is used for disease diagnosis, drug development, and patient data management in health matters.In the automotive sector, autonomous cars use AI technology to improve road safety and change how we travel.On the other hand, AI enables more sophisticated data analysis, better risk management, and more intelligent investment decision-making.Artificial intelligence continues to develop rapidly and plays a central role in changing our daily lives and various industries' operations.

AI Agent-Based
Agent-Based Modelling and Simulation (ABMS) is a multi-agent system concept applied to a simulation model's basic structure [6].These agents have behaviours that can be described through simple rules, and they interact, influencing each other's behaviour.With the ability to represent any system element, ABMS enables in-depth analysis of the dynamics of interactions between agents, opening the door to a better understanding of the behaviour of complex systems.This approach provides a more adaptive and realistic way to model and simulate systems that involve many interactions and variability [7].Simulation using ABMS is one of the most effective ways to predict people's behaviour in emergency conditions.AI Agent Base technology enables the creation of highly realistic simulation models, predicting fire behaviour, smoke spread, and human response in emergencies.With precise data and advanced algorithms, AI Agent Base has the potential to change the way building designers think about and design security systems, resulting in safer buildings, minimising disaster risks, and improving occupant safety by providing deeper insights into disaster planning and evacuation strategies.

Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application (Irfan Yusuf Muttaqien
AI Agent Base simulation has tremendous potential to improve building security against disasters.AI's ability to test various evacuation scenarios enables early identification of potential risks, a proactive step in building safety planning.The results of these simulations allow building designers to design more resilient structures for disasters, install more intelligent and responsive warning systems, and organise more efficient evacuation routes.Building occupants can feel safer and more protected, while the risk of loss in disaster scenarios can be minimised.In this way, AI Agent Base technology has a crucial role in advancing efforts to improve the safety and sustainability of our environment. Using AI Agent Base technology in emergency simulation opens opportunities for significant improvements in building security against potential disaster threats.Through the implementation of artificial intelligence (AI), early detection, forecasting the development of disaster impacts, and real-time monitoring can be carried out, ensuring efficient and safe evacuation routes.This technology allows training security personnel more realistically, creating a simulated experience close to actual conditions.

AnyLogic
AnyLogic is one platform that utilises the concept of Agent-Based Modeling (ABM) in simulation development.With the Agent-Based concept, active entities, known as agents, must be identified and their behaviour determined.They may be people, households, vehicles, appliances, products, companies, or whatever is relevant to the system.The connection between them is established, environment variables are established, and simulations are executed.The global dynamics of the system then arise from the interaction of many individual behaviours.The AnyLogic model enables analysts, engineers, and managers to gain deeper insights and optimise complex systems and processes across various industries.
In an emergency evacuation simulation using AnyLogic, everyone within the building can be represented as a unique agent.They can be set up to respond to disasters based on careful programming so that simulations can illustrate a variety of possible evacuation scenarios.Evacuation scenarios include the arrangement of evacuation routes, an understanding of the different levels of danger, and how individuals interact with each other during the evacuation process.

AnyLogic Usage Steps
The steps to use the Anylogic application are explained as follows.First, install AnyLogic from the official website and create a new project.Choose the type of project that fits the purpose, such as entitybased, agent-based, or dynamics-based simulation.Next, develop our model by importing graphical components corresponding to the modelling process, such as entities, processes, resources, and queues.Set the parameters, variables, and rules required for the model.Define initial scenarios and conditions.Once the model is complete, run simulations and analyse the results using graphs, reports, and statistics provided by AnyLogic.You can experiment with changing parameters and conditions to answer specific questions or identify the best solution.Then, the model will be documented, and an analysis will be performed to complete the simulation.There are three main stages in the modelling process when using AnyLogic for emergency evacuation simulation: 1.A physical model is built to match the layout of the actual simulated environment.AnyLogic has a scale that can be adapted to the user's needs, and AnyLogic's pedestrian displacement model has appropriate spatial markers to help draw graphics.

RESULTS AND DISCUSSION
Table 1 shows the literature researchers collected on using analogies in disaster evacuation.The data above shows that AnyLogic is crucial in optimising emergency evacuation strategies in various contexts, such as field stations, complex spatial environments, building plans, and crises such as gas leaks.AnyLogic is used as an efficient simulation tool to evaluate and improve evacuation plans, improve visitor safety, and

Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application (Irfan Yusuf Muttaqien
provide optimal solutions regarding evacuation time.AnyLogic has also proven effective in developing simulation models for mass evacuation strategies, predicting various situations, and improving evacuation efficiency at subway stations, cancer medical centres, and airports.AnyLogic can model the safest and optimal emergency evacuation plan, improve the efficiency of emergency evacuation, and evaluate the efficiency and safety of disaster evacuation routes.AnyLogic can model individual behaviour in fire evacuation situations and enrich understanding of evacuation dynamics in real situations.Factors affecting evacuation efficiency include the number of people, density, and movement speed.Potential improvements in the safety and accessibility of evacuation routes involve increasing the number of emergency exits, adding emergency stairs, and improving lighting quality.

Advantages of Using Anylogic
AnyLogic can determine evacuation routes efficiently through agent-based simulation models.AnyLogic identifies the most optimal evacuation route in various emergencies at emergency sites.The main advantage of an agent-based AnyLogic lies in its ability to model individual behaviour, allowing everyone to have an agent who can make independent decisions about evacuation routes.AnyLogic libraries, such as Pedestrian, Agent-based, and Space Markup, provide powerful tools for building models of complex spatial environments.With the social style model, AnyLogic creates real-world behavioural simulations, enabling evaluation of the dynamics of individual interactions in evacuation and designing more efficient evacuation strategies in hotel emergencies.AnyLogic could determine individual behaviour in simulations and monitor them to analyse the time taken by each agent to reach evacuation routes and safe points.In the analysis results, the presence and location of firefighting equipment, emergency stairs, and the placement of evacuation signs significantly affect the evacuation time.This data is critical to analysing the efficiency of evacuation routes and identifying areas where congestion points are likely.Leveraging advanced modelling and analysis features, AnyLogic provides an in-depth understanding of evacuation dynamics, enabling breakthrough improvements to improve emergency response and safety.For example, AnyLogic helps researchers create simulation models that visualise workflows in the Emergency Department (ED) and map real-time data components into these models.Thus, AnyLogic can assist in understanding ED operational processes and how real-time data interacts with these processes, as demonstrated in the process flow diagram (Figure 1) mentioned in the diagram.

Evaluation of Emergency Access Evacuation Routes Using Agent-Based Model Application (Irfan Yusuf Muttaqien
AnyLogic uses analysis and simulation to evaluate the safety of evacuation routes by considering factors such as overcrowding and physical constraints, such as locked or blocked doors.The evaluation results show potential improvements in ensuring that evacuation routes are always accessible and safe to use.By utilising AnyLogic software, managers can build accurate simulation models to develop effective and efficient emergency evacuation plans to minimise the risk of accidents and ensure user safety.AnyLogic analysis or similar simulation tools are an essential foundation for improving the efficiency of evacuation points in future disaster planning.By collecting and analysing geographic and demographic data and simulating various disaster scenarios, we can identify optimal evacuation point locations, design efficient evacuation routes, and prepare communities with better training and understanding of evacuation procedures.The results of this analysis become the basis for better planning and improved safety when facing disaster situations.

CONCLUSION
Based on the analysis, AnyLogic optimises emergency evacuation strategies in various contexts.Anylogic has the advantage of determining the most efficient evacuation route, determining evacuation time, and being able to evaluate evacuation routes in certain buildings.With the combination of these features, AnyLogic becomes an efficient tool in disaster planning and safety.The application of AnyLogic provides optimal solutions for evacuation times, improves visitor safety, and is effective in various facilities, such as subway stations, cancer medical centres, and airports.
Facility managers and evacuation system designers need to deeply understand the capabilities and features of AnyLogic to maximise its use in future disaster planning.Intensive training on using AnyLogic in various evacuation scenarios can provide the expertise needed to design more effective and responsive evacuation plans.In addition, close collaboration between AnyLogic users, software developers, and security experts can result in continuous updates to the software to continuously adapt to the evolving security and disaster systems needs.The research has not only shed light on the potential of AnyLogic in optimising emergency evacuation strategies.Still, it has also emphasised the importance of leveraging AI technology to enhance building safety measures and improve disaster response planning.

Table 1 .
Literature about Agent-Based Artificial Intelligence AnyLogic.