Design and Development of an Agent-based Social Simulation Visualization Tool for Indoor Crowd Analytics based on the library Three.js

Pablo Aznar. (2017). Design and Development of an Agent-based Social Simulation Visualization Tool for Indoor Crowd Analytics based on the library Three.js. Trabajo Fin de Titulación (TFG). ETSI Telecomunicación, Universidad Politécnica de Madrid.

Social simulation consists of modeling social behavior with computational methods. These models allow researchers to explain social phenomenons and to analyze how people interact with each other in specific situations. In addition, social simulation provides information that can be interpreted in a lot of ways and have different utilities. For example, it allows to see how people behave and move in the evacuation of a building or to see how its energy consumption varies. All the data obtained from the simulation can be interpreted in order to know which is the best way to design a building according to evacuation needs or to make an estimation of the energy consumption as well as the temperature variation. There are different tools that analyze people behaviors inside buildings in a numerical way. However, there is an increasingly need of visualizing these behaviors in a visual form, that means, in a 3D environment. This will give the opportunity to analyze the results of the simulation in an easier and more intuitive way because it will be simpler and faster to find solutions for possible problems. Thus, being able to visualize in a 3D environment the human behavior will allow, in a very easy and simple way, to obtain very valuable data. Consequently, the objective of this project is to design and to develop a tool that al- lows to visualize a 3D environment in which there are people interacting with each other. This will provide a very useful and helpful way to analyze different situations. For this purpose, Three.js will be used, which is a JavaScript library to visualize and to animate 3D environments. This tool will create a 3D model of a specific building interior with its different objects and in which it will be possible to observe people doing several activities. All this could be very useful because it will give the opportunity to see how different aspects vary depending on human behaviors, as the temperature variation of a room when nobody is in it or when it is plenty of people. Furthermore, other aspect that could be analyzed is the energy consumption, as it could be represented how energy consumption varies according to the state, on or off, of the devices when people are not using them.