Enhancing Crowd Safety and Inclusivity: Fairness-Aware Evacuation Strategies and Surge Prevention Measures for Smart Cities

Belangrijkste concepten

Implementing fair evacuation strategies that consider the diverse needs of all individuals, including vulnerable populations, and developing preventative approaches to minimize the occurrence of crowd surges and enhance overall crowd safety.

Samenvatting

This paper proposes two key methodologies to address the limitations of current crowd management practices in smart cities:

Fair Evacuation Strategies:

Introduces a "Normalized Evacuation Time Disparity" (NETD) metric to evaluate the fairness of evacuation plans, aiming to ensure similar evacuation times for vulnerable and healthy individuals.
Simulates various crowd scenarios and compares three evacuation strategies: Randomly Assigned Gates (RGA), Vulnerable Exclusive Gate Assignment (VEGA), and Closest Gate Assignment (CGA).
The results show that VEGA can improve fairness by 78% on average compared to RGA and CGA in scenarios where vulnerable individuals are concentrated near a single exit.
However, CGA demonstrates better fairness in scenarios where the crowd is centered or evenly dispersed, improving fairness by 21.5% on average compared to RGA and VEGA.

Preventative Surge Mitigation:

Proposes a preventative approach involving the adjustment of attraction locations and switching between stage performances at large-crowded events.
Introduces metrics like Panic State, Surge State, and Crowded State to dynamically assess and manage crowd conditions.
Utilizes an agent-based simulation tool in NetLogo to evaluate the effectiveness of the prevention strategy.
The results show that increasing the distance between stages can reduce the frequency of stage switching by 26% and the average panic/surge by 34% on average.
The optimal timing for stage switching, captured by the Switch Index (SI), is crucial in balancing the frequency of disruptions and the average panic/surge.
The paper emphasizes the need for adaptive and inclusive crowd management strategies that consider the diverse needs of all individuals to enhance safety and fairness in smart city environments.

Statistieken

"When the number exceeds 6 per square meter, the limited available space forces tight packing and diminishes individual control, significantly increasing the likelihood of a surge."
"On October 29, 2022, a Halloween event occurred in Seoul, South Korea, attracting tens of thousands of costumed attendees to the Itaewon district. This marked the first unrestricted Halloween celebration in over two years due to COVID-19 lockdowns. The massive crowd in the narrow streets, coupled with limited entry and exit points, created a dangerous situation. This catastrophe led to one of South Korea's worst stampede disasters, with 156 deaths and 170 crush injuries."
"The average running speed of healthy and young individuals (aged 20-45 years) is ≈5.4 miles per hour or ≈2.4 meters per second. Vulnerable people, such as the elderly, move at a slower pace as the speed of humans decreases by 20% every decade."

Citaten

"Instances of casualties resulting from large crowds persist, highlighting the existing limitations of current crowd management practices in Smart Cities."
"One notable drawback is the insufficient provision for disadvantaged individuals who may require additional time to evacuate due to their slower running speed."
"The sheer number of people in a confined space can create a dangerous situation that can quickly spiral out of control, resulting in stampedes and crush injuries."

Belangrijkste Inzichten Gedestilleerd Uit

Towards Fairness-aware Crowd Management System and Surge Prevention in Smart Cities

by Yixin Zhang,... om arxiv.org 04-24-2024

https://arxiv.org/pdf/2311.02228.pdf

Diepere vragen

How can the proposed fairness-aware evacuation strategies be further improved to account for social dynamics, such as the tendency of groups (families, friends) to select the same escape route during an emergency?

In order to enhance the proposed fairness-aware evacuation strategies to consider social dynamics like the tendency of groups to choose the same escape route, several adjustments can be made:

Group Identification: Implement a system that allows groups to register together before an event. This way, organizers can identify and track groups during an evacuation, ensuring they stick together and follow the designated route.

Group-Specific Notifications: Provide personalized notifications to groups, guiding them to specific exits based on their registration information. This can prevent groups from getting separated and ensure they follow a safe path.

Dynamic Route Allocation: Develop algorithms that dynamically assign evacuation routes based on the distribution of groups. By considering the locations of different groups, the system can optimize routes to prevent congestion and ensure efficient evacuation.

Group Leader System: Designate group leaders within each group who are responsible for ensuring everyone in their group follows the evacuation plan. These leaders can be equipped with communication devices to receive real-time updates and instructions.

Interactive Maps: Provide interactive maps or digital displays showing the location of each group and the recommended route. This visual aid can help groups navigate effectively and stay together during the evacuation.

Training and Drills: Conduct training sessions and evacuation drills specifically tailored for groups. This can familiarize them with the evacuation process, designated routes, and the importance of sticking together during emergencies.

By incorporating these enhancements, the fairness-aware evacuation strategies can better accommodate social dynamics and ensure the safety of groups during emergency evacuations.

How can the preventative approach be enhanced to address the challenge of individuals who may not adhere to the recommended guidelines set by the organizers or the automated crowd management system?

To address the challenge of individuals who may not comply with recommended guidelines during crowd management, the preventative approach can be strengthened through the following strategies:

Behavioral Nudges: Implement behavioral nudges through the crowd management system to encourage individuals to follow guidelines. This can include visual cues, announcements, or notifications that remind people to adhere to safety protocols.

Incentive Systems: Introduce incentive systems that reward individuals for following guidelines. This could involve gamification elements, such as points or rewards for compliance, to motivate better behavior.

Peer Influence: Leverage social influence by highlighting the majority of people who are adhering to guidelines. This can create social pressure for non-compliant individuals to conform to the expected behavior.

Real-Time Feedback: Provide real-time feedback to individuals who are not following guidelines, alerting them to their behavior and its impact on crowd safety. This feedback can prompt immediate course correction.

Crowd Monitoring Technologies: Utilize advanced technologies like AI-powered video analytics to identify non-compliant behavior in real-time. Security personnel can then intervene and guide individuals to adhere to the rules.

Emergency Response Teams: Deploy dedicated emergency response teams trained to handle non-compliant individuals effectively. These teams can engage with and guide non-compliant individuals to ensure their safety and the safety of others.

By incorporating these enhancements, the preventative approach can better address the challenge of individuals who may not adhere to guidelines, ultimately improving crowd safety and management effectiveness.

What other technological advancements, beyond sensors and wearables, could be leveraged to enhance the real-time monitoring and decision-making capabilities of crowd management systems in smart cities?

In addition to sensors and wearables, several other technological advancements can be leveraged to enhance real-time monitoring and decision-making capabilities in crowd management systems within smart cities:

Artificial Intelligence (AI) and Machine Learning: AI algorithms can analyze crowd behavior patterns, detect anomalies, and predict potential issues in real-time. Machine learning models can optimize evacuation routes based on dynamic crowd data.

Drones: Drones equipped with cameras and sensors can provide aerial views of crowd movements, helping authorities monitor and manage crowds more effectively. They can also be used for rapid response and communication during emergencies.

Blockchain Technology: Blockchain can enhance the security and transparency of crowd management data. It can be used to securely store and share information, such as evacuation plans, crowd density, and emergency protocols.

Augmented Reality (AR) and Virtual Reality (VR): AR and VR technologies can create immersive simulations for training purposes and real-time visualization of crowd dynamics. This can aid decision-makers in understanding and responding to crowd situations more effectively.

Predictive Analytics: By analyzing historical data and real-time information, predictive analytics can forecast crowd behavior, identify potential risks, and suggest proactive measures to prevent incidents.

Internet of Things (IoT): IoT devices can be deployed to collect data on crowd density, temperature, and air quality in real-time. This data can be used to make informed decisions and optimize crowd management strategies.

Mobile Applications: Mobile apps can serve as communication tools during emergencies, providing real-time alerts, evacuation instructions, and updates to individuals in crowded areas.

By integrating these technological advancements into crowd management systems, smart cities can enhance their monitoring capabilities, improve decision-making processes, and ensure the safety and well-being of individuals during mass gatherings and emergencies.