Banos / Lang / Marilleau Agent-Based Spatial Simulation with NetLogo Volume 1

1. Auflage 2015
ISBN: 978-0-08-100723-5
Verlag: Morgan Kaufmann
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 278 Seiten, Format (B × H): 152 mm x 229 mm

Agent-based modeling is a flexible and intuitive approach that is close to both data and theories, which gives it a special position in the majority of scientific communities. Agent models are as much tools of understanding, exploration and adaptation as they are media for interdisciplinary exchange. It is in this kind of framework that this book is situated, beginning with agent-based modeling of spatialized phenomena with a methodological and practical orientation. Through a governing example, taking inspiration from a real problem in epidemiology, this book proposes, with pedagogy and economy, a guide to good practices of agent modeling. The reader will thus be able to understand and put the modeling into practice and acquire a certain amount of autonomy.



Featuring the following well-known techniques and tools:
- Modeling, such as UML
- Simulation, such as the NetLogo platform
- Exploration methods
- Adaptation using participative simulation

Zielgruppe

Advanced practitioners from industry and academia, and graduate students in the field of multi-agent systems

Autoren/Hrsg.

Banos, Arnaud

Lang, Christophe

Marilleau, Nicolas

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz

Inhaltsverzeichnis

1. Introduction to the Agent Approach 2. Description Formalisms in Agent Models 3. Introduction to NetLogo 4. Agent-Based Model Exploaration 5. Dynamical Systems with NetLogo 6. How to Involve Stakeholders in the Modeling Process

Leseproben

2 Description Formalisms in Agent Models
Fabrice Bouquet; David Sheeren; Nicolas Becu; Benoît Gaudou; Christophe Lang; Nicolas Marilleau; Claude Monteil Abstract
This chapter will aim to present good practice in, and the benefits of, formalization in modeling multiagent systems (MAS). To achieve this, the authors will first reiterate the usefulness of modeling systems, while placing the paradigms associated with a multiagent approach in context. Then, they will argue that the use of graphic modeling languages enhances the exchanges between the parties involved in the design of an MAS. Following this, two types of graphic models based on the same semantic base are presented: Unified Modeling Language (UML) and Agent Modeling Language (AML). The first graphic model is intended for general use and facilitates its users to analyze the ontology and dynamics of the modeled system. The second graphic model uses paradigms specific to agents and facilitates its users to create a design which is closer to the MAS which will be produced. After having discussed the relative merits of each of these graphic model types and presented some possible extensions, the chapter discusses the utility of, and a method for, documenting a multiagent model. In order to do this, the Overview, Design concepts, Details (ODD) protocol, which guides the modeler in the creation of a documentation of the objectives, constitutive elements and specific properties of the model, is presented. Keywords Agent Modeling Language (AML) Coherence verification Documentation Formalization Interaction model Object Constraint Language (OCL) Overview, design concepts, details (ODD) protocol Static diagrams Unified Modeling Language (UML) 2.1 Introduction
This chapter will aim to present good practice in, and the benefits of, formalization in modeling multiagent systems (MAS). To achieve this, the authors will first reiterate the usefulness of modeling systems, while placing the paradigms associated with a multiagent approach in context. Then, they will argue that the use of graphic modeling languages enhances the exchanges between the parties involved in the design of an MAS. Following this, two types of graphic models based on the same semantic base are presented: Unified Modeling Language (UML) and Agent Modeling Language (AML). The first graphic model is intended for general use and facilitates its users to analyze the ontology and dynamics of the modeled system. The second graphic model uses paradigms specific to agents and facilitates its users to create a design which is closer to the MAS which will be produced. After having discussed the relative merits of each of these graphic model types and presented some possible extensions, the chapter discusses the utility of, and a method for, documenting a multiagent model. In order to do this, the Overview, Design concepts, Details (ODD) protocol, which guides the modeler in the creation of a documentation of the objectives, constitutive elements and specific properties of the model, is presented. We illustrate each of the concepts presented (UML, AML and ODD) through their application to an example which will be a recurrent theme in the remaining of this chapter. 2.2 Recurrent example
Many applications exist which have clearly demonstrated the utility of the agent approach for modeling complex phenomena. These involve numerous domains, such as ecology, social science or epidemiology. We have chosen to address the domain of epidemiology because, first, this theme takes in several other domains, in particular ecology and social sciences, and second, a wide range of multiagent concepts can be involved in modeling complex phenomena such as these. As such, this chapter, and those that follow it, will be thematically linked through a recurrent modeling example which is based on an epidemiological phenomenon. The phenomenon in question is the geographic dispersion of an epidemic transmitted to humans by mosquitoes. The aim of the model is to understand and measure the impact of the pendular journeys people take (moving from home ? work) on the development and spread of a contagious disease. We have decided to study malaria, which is present in many African countries, and we apply this to the Maroua subregion of Cameroon (see Figure 2.1). Figure 2.1 Map of the Maroua subregion (Cameroon) The only way in which malaria is transmitted to humans is through the bite of an infected anopheles mosquito. A healthy mosquito becomes infected when it bites an infected person. People cannot pass the disease on to other people, and mosquitoes cannot pass the disease on to other mosquitoes. By itself, the mosquito has a very small movement radius, of approximately 50 m a day, but it is present throughout the territory. However, the real vector for the spread of the disease seems to be people because they need to travel over large distances to conduct their daily activities. Most of the inhabitants of the Maroua live in urbanized areas (black in Figure 2.1). The crop farmers of the region travel in a pendular movement between their homes (in town) and the area that they farm, outside the town (white and gray areas in Figure 2.1). The wetland zones correspond to wet areas where the mosquitoes can lay their eggs. In the following, this system will be modeled using UML (section 2.3.1), and then AML (section 2.3.2), and will be documented using the ODD protocol (section 2.4). In this way, the readers will discover the different forms of model description through an example of each. 2.3 Formalization of agent models
In this section, we will first detail the reasons for formalizing MAS, and then we will present two tools designed to aid this formalization – UML and AML – which we will illustrate using our recurring example. Formalization is a process which has three goals. The first goal is related to the system under study. A tool is required that is suitable for gaining an understanding of the system concerned. The second goal is connected with abstraction, which helps us to not be restricted by technical considerations linked to the simulation components. The third goal is the generation of a code, which will make it possible to transition from model to implementation. In addition, graphic formalizations, such as UML and AML, also allow us to streamline communication about the content of the model among several people. These languages have the advantage of being able to synthetically show, using one or several diagrams, complex mechanisms and structures. The graphic nature of these diagrams simplifies their understanding for non-programmers. In this chapter, we have chosen to concentrate on UML and AML; the latter is an extension of the former, and it is more specifically dedicated to the agent paradigm through its formalization of the description of agent behaviors and interactions. The various diagrams developed from UML may also, of course, be used for model development. However, its generic nature causes certain problems when it comes to using it within a specific context. Thus, for spatialized simulations, it is necessary to use a language which can represent space and its constraints, such as the representations underlying Geographic Information Systems (GIS) [CHI 13]. 2.3.1 UML
In the 1970s and 1980s, there was disagreement between those who believed in modeling data and those who believed in functional modeling. In this period, the use of flow and relational diagrams was generally considered to be mutually exclusive. These two camps finally came to an agreement at around the end of the 1980s, and realized that most projects could benefit from the use of both model types. This reconciliation was followed by the emergence of numerous object-oriented analysis and modeling methods. However, each method had its own specific notation and definition of terms such as object, type and class. There was no common standard. The number of modeling languages increased from less than 10 to more than 50 between 1989 and 1994. At the end of 1994, Grady Booch and Jim Rumbaugh announced their collaboration on the development of a Unified Method. They were later joined by Ivor Jacobson. In the end, after several years of experimentation with various notations and concepts, the group established a semantics for object-oriented concepts and agreed on a common notation on the basis of several of the notations and concepts with which they had experimented. During 1996, the Unified Method developed into the UML. This new name was designed to emphasize the fact that UML was a modeling language and not a method. Its aim was to provide an expressive notation to define a semantics for implied concepts, and to leave the development process choice open. In the end, UML, developed from the combination of the three methods of object modeling, Object Modeling Technique (OMT), Booch and Object Oriented Software Engineering (OOSE), became an essential standard. Originally created to enable a developer to represent, specify, analyze and visualize the structure of a project in object-oriented programming, UML is today used in a large number of fields. The development of UML is quite like the development of software in that there are major versions as well as...