AnyLogic is a simulation tool with over 15,000 users in 60 countries. The software product is intended for designing and optimizing business processes or any complex systems, such as a production shop, an airport, a hospital, etc. The tool supports all methods of business modeling - system dynamics, discrete-event (process) and agent-based modeling. The main emphasis in the development of the product is on its flexibility and ease of use for users inexperienced in creating models. Among the users are large international companies, government agencies, international alliances: NATO, NASA, Boeing, IBM, General Motors, HSBS, PWC, Accenture, Jonson&Jonson.

2014: AnyLogic 7

1990-1998

In the early 1990s, there was a great interest in computer science in constructing a mathematically interpretable description of the interaction of parallel processes. What affected the approaches to the analysis of the correctness of parallel and distributed programs. A group of scientists from the St. Petersburg Polytechnic University has developed software for analyzing the correctness of the system; the new tool was called COVERS (Parallel Verification and Simulation). The analyzed system of processes was specified graphically, by describing its structure and the behavior of individual parallel components that could interact with the environment - with other processes and the environment. The tool has been used in research projects at Hewlett-Packard.

In 1998, the success of this research inspired the laboratory to start a commercial company with the mission of creating new simulation software. The emphasis in the development was placed on applied methods: modeling of stochastic systems, optimization and visualization of the model. The new software, released in 2000, was based on the latest advances in information technology: object-oriented approach, elements of the UML standard, Java programming language, modern GUI, etc.

The product was named AnyLogic because it supported all three known simulation methods:

• system dynamics;
• discrete-event modeling|discrete-event (process) modeling;
• agent modeling.

As well as any combination of these approaches within the same model. The first version was assigned index 4 - Anylogic 4.0, since the numbering continued the version history of the previous development - COVERS 3.0.

A huge step forward was made in 2003 with the release of AnyLogic 5 focused on business modeling. With the help of AnyLogic, it became possible to develop models in the following areas:

AnyLogic & Java

AnyLogic includes a graphical modeling language and also allows the user to extend the created models using the Java language. The integration of the Java compiler into AnyLogic provides more opportunities for creating models, as well as the creation of Java applets that can be opened by any browser. These applets make it easy to place AnyLogic models on websites. In addition to Java applets, AnyLogic Professional supports the creation of Java applications, in which case the user can run the model without installing AnyLogic.

Simulation methods

AnyLogic models can be based on any of the major simulation modeling paradigms: discrete event modeling, system dynamics, and agent-based modeling.

System dynamics and discrete-event (process) modeling, by which we mean any development of GPSS ideas, are traditional, well-established approaches, agent-based modeling is relatively new. System dynamics operates mainly with processes that are continuous in time, while discrete-event and agent-based modeling - with discrete ones.

System dynamics and discrete event modeling have historically been taught to completely different groups of students: management, production engineers, and control system design engineers. As a result, three different almost non-overlapping communities have emerged, which almost never communicate with each other.

Agent-based modeling has been a strictly academic field until recently. However, the growing demand for global optimization on the part of business has forced leading analysts to pay attention to agent-based modeling and its combination with traditional approaches in order to obtain a more complete picture of the interaction of complex processes of various nature. Thus, the demand for software platforms was born that allows integrating different approaches.

Now let's consider simulation modeling approaches on the abstraction level scale. System dynamics, by replacing individual objects with their aggregates, assumes the highest level of abstraction. Discrete event simulation works in the low and medium range. As for agent-based modeling, it can be applied at almost any level and on any scale. Agents can represent pedestrians, cars, or robots in a physical space, a customer or salesperson at a mid-level, or competing companies at a high level.

When developing models in AnyLogic, you can use concepts and tools from several modeling methods, for example, in an agent-based model, use the methods of system dynamics to represent changes in the state of the environment, or in a continuous model of a dynamic system, take into account discrete events. For example, supply chain management using simulation modeling requires the description of supply chain participants by agents: manufacturers, sellers, consumers, a network of warehouses. At the same time, production is described within the framework of discrete-event (process) modeling, where the product or its parts are applications, and cars, trains, stackers are resources. The deliveries themselves are represented by discrete events, but the demand for goods can be described by a continuous system-dynamic diagram. The ability to mix approaches allows you to describe the processes of real life, and not to adjust the process to the available mathematical apparatus.

Simulation environment

The modeling environment also includes: low-level modeling constructs (variables, equations, parameters, events, etc.), representation forms (lines, squares, ovals, etc.), analysis elements (databases, histograms, graphs), standard pictures and forms of experiments.

The AnyLogic modeling environment supports the design, development, documentation of the model, the execution of computer experiments with the model, including various types of analysis - from sensitivity analysis to optimization of model parameters relative to some criterion.

Libraries AnyLogic

AnyLogic includes a set of the following standard libraries:

• Enterprise Library designed to support discrete event modeling in areas such as Manufacturing, Supply Chain, Logistics and Healthcare. Using the Enterprise Library, you can model real world systems in terms of entities (transactions, customers, products, vehicles, etc.), processes (sequences, queues, delays), and resources. The processes are defined in the form of a block diagram.
• Pedestrian Library created to simulate pedestrian flows in a "physical" environment. This allows you to create models with a lot of pedestrian traffic (like metro stations, security checks, streets, etc.). The models support the accounting of traffic density statistics in various areas. It guarantees acceptable operation of facilities with occupancy restrictions, evaluates downtime in certain areas, and detects potential problems with internal geometry - such as the effect of adding too many obstacles - and other phenomena. In models created with the Pedestrian Library, pedestrians move continuously, reacting to different kinds of obstacles (walls, different kinds of areas) in the same way as ordinary pedestrians. Pedestrians are modeled as interacting agents with complex behavior. To quickly describe pedestrian flows, the Pedestrian Library provides a high-level box diagram interface.
• Rail Yard Library supports modeling, simulation and visualization of marshalling yard operations of any complexity and scale. Marshalling yard models can use combined modeling methods (discrete-event and agent-based modeling) related to transportation activities: loading and unloading, resource allocation, maintenance, various business processes.

Notes

Literature

• Karpov, Yu. G. Simulation modeling of systems. Introduction to modeling with AnyLogic 5. - St. Petersburg: BHV-Petersburg, 2006. - 400 p. -

Federal Agency for Education Ural State Technical University – UPI

named after the first President of Russia B.N. Yeltsin

M. V. Kiseleva

SIMULATION MODELING OF SYSTEMS IN THE ANYLOGIC ENVIRONMENT

Teaching aid

Scientific editor professor, doctor of technical sciences L. G. Dorosinsky

Published by decision of the editorial and publishing council of USTU-UPI dated May 26, 2009.

Yekaterinburg USTU-UPI

UDC 004.434:004.94(075.8) LBC 32.973.26-018.2а73

Reviewers:

Department of Automation and Information Technologies - Ural State College named after I. I. Polzunov (Associate Professor, Candidate of Technical Sciences V. V. Kiyko);

S. N. Kiselev, gen. director of CJSC Ural Regional Intersectoral Center for Scientific and Technical Development.

Kiseleva M.V.

K44 Simulation modeling of systems in the AnyLogic environment: teaching aid / M. V. Kiseleva. Yekaterinburg: USTU - UPI,

The tutorial is designed to study the methods and tools for building simulation models in the AnyLogic tool environment.

Contains three laboratory works on creating simulation models - discrete-event, system-dynamic and agent-based. For each model, a detailed statement of the problem is given, the structure and implementation of the model in the AnyLogic environment are analyzed. Techniques of visual visualization of the process under study, interactive animation with the possibility of changing the system parameters during the process modeling are studied.

Bibliography: 4 titles. Rice. 67.

UDC 004.434:004.94(075.8)

BBK 32.973.26-018.2ya73

© USTU-UPI, 2009

© Kiseleva M. V., 2009

INTRODUCTION Simulation modeling

Modeling is a problem solving method in which the system under study is replaced by a simpler object that describes the real system and is called a model.

Simulation is used in cases where it is impossible or impractical to conduct experiments on a real system, for example, due to the high cost or duration of the experiment in real time.

There are physical and mathematical modeling. An example of a physical model is a reduced copy of an aircraft blown in an air stream. When using mathematical modeling, the behavior of the system is described using formulas. Simulation models are a special kind of mathematical models.

A simulation model is a computer program that describes the structure and behavior of a real system over time. The simulation model allows obtaining detailed statistics on various aspects of the system functioning depending on the input data.

Simulation modeling is the development of computer models and the setting of experiments on them. Ultimately, the purpose of modeling is to make reasonable, expedient management decisions. Computer modeling is now becoming an obligatory stage in making responsible decisions in all areas of human activity due to the complication of systems in which a person must act and which he must manage. Knowledge of the principles and capabilities of simulation modeling, the ability to create and apply models are essential requirements for an engineer, manager, business analyst.

Purpose and capabilities of the AnyLogic tool environment

Modern modeling systems support the entire arsenal of the latest information technologies, including advanced graphical shells for the purpose of constructing models and interpreting the output results of modeling, multimedia tools, real-time animation, object-oriented programming, Internet solutions, etc. This manual describes methods and techniques for building models using the AnyLogic tool system.

The AnyLogic package is a domestic professional tool of a new generation, which is designed for the development and research of simulation models. Product developer - XJ Technologies, St. Petersburg; email address: www.xjtek.ru .

AnyLogic was developed on the basis of new ideas in the field of information technology, the theory of parallel interacting processes and the theory of hybrid systems. Thanks to these ideas, the construction of complex simulation models is extremely simplified, it is possible to use one tool when studying various modeling styles.

The AnyLogic software tool is based on an object-oriented concept. Another basic concept is to represent a model as a set of interacting, parallel activities. An active object in AnyLogic is an object with its own functioning that interacts with the environment. It can include any number of instances of other active objects.

The graphical modeling environment supports the design, development, documentation of the model, the execution of computer experiments, and the optimization of parameters relative to some criterion.

When developing a model, you can use elements of visual graphics: state diagrams (statecharts), signals, events (timers), ports, etc.; synchronous and asynchronous event scheduling; asset libraries

objects.

A user-friendly interface and numerous support tools for developing models in AnyLogic make not only the use, but also the creation of computer simulation models in this modeling environment accessible even for beginners.

When developing a model on AnyLogic, you can use concepts and tools from several classical areas of simulation modeling: dynamic systems, discrete event modeling, system dynamics, agent-based modeling. In addition, AnyLogic allows you to integrate various approaches in order to get a more complete picture of the interaction of complex processes of various nature.

This manual describes three simulation models: discrete-event, system-dynamic and agent-based. For each model, a detailed statement of the problem is given, the structure of the model is analyzed, the process of building the model in the AnyLogic environment is described, and its behavior is studied.