Daniel Formolo

An agent-based evacuation model with social contagion mechanisms and cultural factors

A fire incident at a transport hub can cost many lives. To save lives, effective crisis management and prevention measures need to be taken. In this project, the effect of cultural factors in managing and preventing emergencies in public transport systems is analysed. An agent–based model of an evacuating crowd was created. Socio-cultural factors that were modelled are: familiarity with environment, response time and social contagion of fear and beliefs about the situation. Simulation results show that (1) familiarity and social contagion decrease evacuation time, while increasing the number of falls; (2) crowd density and social contagion increase evacuation time and falls. All three factors show different effects on the response time. This model will be used by transport operators to estimate the effect of these socio-cultural factors and prepare for emergencies.

Simulating crowd evacuation with socio-cultural, cognitive, and emotional elements

In this research, the effects of culture, cognitions, and emotions on crisis management and prevention are analysed. An agent-based crowd evacuation simulation model was created, named IMPACT, to study the evacuation process from a transport hub. To extend previous research, various socio-cultural, cognitive, and emotional factors were modelled, including: language, gender, familiarity with the environment, emotional contagion, prosocial behaviour, falls, group decision making, and compliance. The IMPACT model was validated against data from an evacuation drill using the existing EXODUS evacuation model. Results show that on all measures, the IMPACT model is within or close to the prescribed boundaries, thereby establishing its validity. Structured simulations with the validated model revealed important findings, including: the effect of doors as bottlenecks, social contagion speeding up evacuation time, falling behaviour not affecting evacuation time significantly, and travelling in groups being more beneficial for evacuation time than travelling alone. This research has important practical applications for crowd management professionals, including transport hub operators, first responders, and risk assessors.

An Adaptive Simulation Tool for Evacuation Scenarios

Building useful and efficient models and tools for a varied audience, such as evacuation simulators for scientists, engineers and crisis managers, can be tricky. Even good models can fail in providing information when the user’s tools for the model are scarce of resources. The aim of this work is to propose a new tool that covers the most required features in evacuation scenarios. This paper starts with a review of current software, prototypes and models simulating evacuation scenarios, by discussing their required and desired features. Based on this overview, we propose our simulator comparing it with other models and commercial tools. Moreover, we discuss the importance of building simulators that cover the minimum requirements to avoid the risk of building inefficient models or tools that do not provide enough insights for users to take right decisions in terms of security policies in crowded events. The implications of this work are to present a new simulation tool and to start a discussion in this research field on mandatory features of evacuation simulation tools that will provide valuable information to users and to find out what the criteria are to define these features.

Human vs. computer performance in voice-based recognition of interpersonal stance

This paper presents an algorithm to automatically detect interpersonal stance in vocal signals. The focus is on two stances (referred to as ‘Dominant’ and ‘Empathic’) that play a crucial role in aggression de-escalation. To develop the algorithm, first a database was created with more than 1000 samples from 8 speakers from different countries. In addition to creating the algorithm, a detailed analysis of the samples was performed, in an attempt to relate interpersonal stance to emotional state. Finally, by means of an experiment via Mechanical Turk, the performance of the algorithm was compared with the performance of human beings. The resulting algorithm provides a useful basis to develop computer-based support for interpersonal skills training.

Studying the Impact of Trained Staff on Evacuation Scenarios by Agent-Based Simulation.

Human evacuation experiments can trigger distress, be unethical and present high costs. As a solution, computer simulations can predict the effectiveness of new emergency management procedures. This paper applies multi-agent simulation to measure the influence of staff members with diverse training levels on evacuation time. A previously developed and validated model was extended with explicit mechanisms to simulate staff members helping people to egress. The majority of parameter settings have been based on empirical data acquired in earlier studies. Therefore, simulation results are expected to be realistic. Results show that staff are more effective in complex environments, especially when trained. Not only specialised security professionals but, especially, regular workers of shopping facilities and offices play a significant role in evacuation processes when adequately trained. These results can inform policy makers and crowd managers on new emergency management procedures.

Extracting Interpersonal Stance from Vocal Signals

The role of emotions and other affective states within Human-Computer Interaction (HCI) is gaining importance. Introducing affect into computer applications typically makes these systems more efficient, effective and enjoyable. This paper presents a model that is able to extract interpersonal stance from vocal signals. To achieve this, a dataset of 3840 sentences spoken by 20 semi-professional actors was built and was used to train and test a model based on Support Vector Machines (SVM). An analysis of the results indicates that there is much variation in the way people express interpersonal stance, which makes it difficult to build a generic model. Instead, the model shows good performance on the individual level (with accuracy above 80%). The implications of these findings for HCI systems are discussed.

Towards Interactive Agents that Infer Emotions from Voice and Context Information

Conversational agents are increasingly being used for training of social skills. One of their most important benefits is their ability to understand the user`s emotions, to be able to provide natural interaction with humans. However, to infer a conversation partner’s emotional state, humans typically make use of contextual information as well. This work proposes an architecture to extract emotions from human voice in combination with the context imprint of a particular situation. With that information, a computer system can achieve a more human-like type of interaction. The architecture presents satisfactory results. The strategy of combining 2 algorithms, one to cover ‘common cases’ and another to cover ‘borderline cases’ significantly reduces the percentage of mistakes in classification. The addition of context information also increases the accuracy in emotion inferences.

A Conversational Agent that Reacts to Vocal Signals

Conversational agents are increasingly being used for training of social skills. One of their most important benefits is their ability to provide natural interaction with humans. This work proposes to extend conversational agents’ benefits for social skills training by analysing the emotion conveyed by the user’s speech. For that, we developed a new system that captures emotions from human voice and, combined with the context of a particular situation, uses this to influence the internal state of the agent and change its behaviour. An example of the system’s use is shown and its limitations and advantages are discussed, together with the internal workflow of the system.