DOCSLIB.ORG

Agent Based Modeling in Anylogic

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Agent Based Modeling in Anylogic Multi-Method Simulation Modeling using AnyLogic Andrei Borshchev, CEO XJ Technologies INFORMS Roundtable Fall Meeting November 3-4, 2007 Seattle © 2007 XJ Technologies www.anylogic.com This presentation… •…is NOT about what you do with the model after it is completed and can be treated as a black box The Optimizer The Model © 2007 XJ Technologies www.xjtek.com 2 This presentation… The Optimizer The Model • …is about HOW you develop simulation models, choose abstraction level and methodology © 2007 XJ Technologies www.xjtek.com 3 Modeling The Real World ? The Problem The Solution © 2007 XJ Technologies www.xjtek.com 4 Modeling The Model The Optimized Model The World of Models The Real World The Problem The Solution © 2007 XJ Technologies www.xjtek.com 5 [Dynamic] Simulation Modeling The Simulation Model • Executable – A set of rules telling how to obtain the next state of the system from the current state • Gives the trajectory of the system in time •Dynamic – Causal and time dependencies – Time-related constraints • Complex – No analytical solution The System © 2007 XJ Technologies www.xjtek.com 6 Methods Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System 1990s Agent Based © 2007 XJ Technologies www.xjtek.com 7 Methods Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System System-level 1990s Agent Based Individual-centric © 2007 XJ Technologies www.xjtek.com 8 Methods Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System 1990s Agent Based Continuous, Discrete, Aggregated Disaggregated © 2007 XJ Technologies www.xjtek.com 9 Abstraction levels. Modeler communities High Abstraction Less Details High aggregation, quantities, global feedback dynamics, … Macro Level SD Strategic Level AB International System Dynamics Conference DE Winter Simulation Conference, INFORMS, IIE Low Abstraction [No business-oriented conference] More Details Micro Level Operational Level Individual objects, exact sizes, distances, velocities, times © 2007 XJ Technologies www.xjtek.com 10 Tools AB SD VenSim PowerSim RePast iThink Swarm ASCAPE NetLogo … DE Arena ED Extend FlexSim PROMODEL Automod SimProcess Simul8 Quest emPlant Witness … © 2007 XJ Technologies www.xjtek.com 11 AnyLogic AB SD DE © 2007 XJ Technologies www.xjtek.com 12 AnyLogic OO Language Java/Eclipse platform Windows, Mac OS, Linux Interactive Animation SD AB Statistics Data presentation OptQuest included + Teamwork support GIS support Open at Java level Export models as applets DE … 1. You can easily vary the level of abstraction and viewpoint until it perfectly fits the problem 2. You can mix different methods in one model © 2007 XJ Technologies www.xjtek.com 13 “No Workarounds!” • Do not fight the modeling language and the tool! • If you feel the SD abstraction level (stocks, flows, feedback dynamics) is enough for the problem, use aggregated view • If the system can be naturally represented as a process (sequence of operations, entities, resources) – use DE • If you are more comfortable with specifying individual behavior of objects (people, vehicles, companies, assets, projects, etc.) – use AB © 2007 XJ Technologies www.xjtek.com 14 Why this is important: cell telecom example Market IT Infrastructure A cell telecom provider runs GPRS in test mode with few customers © 2007 XJ Technologies www.xjtek.com 15 Why this is important: cell telecom example FREE! Market IT Infrastructure Before launching GRPS officially the company declares it is free – as a marketing action © 2007 XJ Technologies www.xjtek.com 16 Why this is important: cell telecom example FREE! Market IT Infrastructure A lot of users start using GPRS because it is free – much more than IT infrastructure can handle © 2007 XJ Technologies www.xjtek.com 17 Why this is important: cell telecom example Market IT Infrastructure Users frustrated with poor QoS leave the company forever © 2007 XJ Technologies www.xjtek.com 18 Why this is important: cell telecom example Model of A combined Market Market + Supply Chain FREE! Model can predict such effects Market IT Infrastructure Model of Supply Chain © 2007 XJ Technologies www.xjtek.com 19 A quick tool demo => © 2007 XJ Technologies www.xjtek.com 20.
Load more

Recommended publications
  • Arena | Anylogic | Flexsim Dr. Alvin Ang
    D R. ALVIN’S PUBLICATIONS A BRIEF COMPARISON OF THREE SIMULATION SOFTWARES ARENA | ANYLOGIC | FLEXSIM DR. ALVIN ANG COPYRIGHTED BY DR ALVIN ANG WWW.ALVINANG.SG CONTENTS Introduction ................................................................................................................ 3 System Dynamics ............................................................................................................... 4 Discrete Event Modeling ..................................................................................................... 5 Agent Based Modeling........................................................................................................ 6 Arena.......................................................................................................................... 7 Made in… ................................................................................................................................................7 Good For… ..............................................................................................................................................7 Bad For… .................................................................................................................................................7 For a trial, ................................................................................................................................................8 Anylogic .....................................................................................................................
    [Show full text]
  • General Dynamics: Capacity Analysis with Anylogic.Pdf
    Success Story: Capacity Analysis with AnyLogic As the West Coast’s largest new construction shipyard, General Dynamics NASSCO is a market leader in the design and construction of ships for the U.S. Navy and commercial customers. Playing a key role in maintaining this strong presence in the U.S. shipbuilding industry is NASSCO’s emphasis on pre-production planning and analysis. With hundreds of thousands of parts flowing through the shipyard each year, however, traditional spreadsheet analysis methods can only provide modest assessment of how those parts, ranging from a hand-sized piece of steel to a fully outfitted deckhouse weighing hundreds of tons, interact with labor, machinery, and workspace to govern NASSCO’s overall shipbuilding throughput and capacity. With AnyLogic simulation software as the centerpiece, NASSCO utilizes a custom-built analysis system called the Large Scale Computer Simulation Modeling System for Shipbuilding (LSMSe) to provide highly detailed and accurate capacity analyses for both current production and potential new work. Developed over a four year period, the LSMSe includes: • A simulation model of the entire shipyard modeled in AnyLogic • Software utilities to define facilities, processes, schedules, labor, and products • Links to product design, scheduling, and business systems software • Optimization software to control parallel scenario analysis over networked computers Implementation of the LSMSe has reduced NASSCO’s recurring analysis labor costs by 75% compared to the former spreadsheet analysis approach with increased detail in the analyses. While significant, an even greater impact is achieved with the ability to identify and eliminate bottlenecks across the integrated shipbuilding system, develop optimum labor assignments across the shipyard, and “right-size” facilities to meet production demand with appropriate capital investments, potentially saving millions of dollars with implementation of the analysis results.
    [Show full text]
  • Anylogic Overview and Roadmap
    overview & roadmap Dr. Andrei Borshchev, CEO, The AnyLogic Company AnyLogic Workshop London May 25 2018 © The AnyLogic Company | www.anylogic.com agenda 1. modeling & dynamic simulation modeling in 2 slides 2. about AnyLogic 3. product and models demo 4. what’s unique about AnyLogic? 5. simulation is migrating to the Cloud 6. AnyLogic Public & Private Cloud 7. AnyLogic roadmap © The AnyLogic Company | www.anylogic.com 2 modeling & dynamic simulation modeling © The AnyLogic Company | www.anylogic.com 3 our favorite slide about modeling Mental The model The solution at the model level Physical Analytical Dynamic (simulation) RISK-FREE SPACE The world of models The real world ? The problem The solution © The AnyLogic Company | www.anylogic.com 4 AnyLogic is a dynamic simulation modeling software • A dynamic simulation model is an “executable”model ̶ A set of rules that allow us to obtain the next state of the system in time from the current state • The model produces a trajectory of the system state in time ̶ Outputs are “observed” as we move Inputs X1 Y X2 Simulation 1 Y2 X3 Model X4 Y4 Y3 © The AnyLogic Company | www.anylogic.com 5 about AnyLogic © The AnyLogic Company | www.anylogic.com 6 AnyLogic is the most popular simulation software LinkedIn user group case studies presented at members as of April 2018 WSC’17 (December 2017) Witness Enterprise Enterprise 2% ExtendSim Automod ProModel Dynamics Dynamics 8% 1% 7% 1% 1% AnyLogic FlexSim AnyLogic AutoMod 2% 3% 30% 33% Simul8 ExtendSim 3% 3% NetLogo Flexsim 8% 8% Repast 5% Simul8 6% Simio
    [Show full text]
  • Software in Support of the System Dynamics Modeling Process
    SILVER: Software in Support of the System Dynamics Modeling Process Nathaniel Osgood Department of Computer Science University of Saskatchewan [email protected] Abstract: While the System Dynamics modeling process can yield invaluable high level insights, it gives rise to a tremendous amount of detail complexity. In the course of their work, modelers must track successive model versions, the motivation for and assumptions underlying particular “what if” scenarios, and the implicit relationships between scenarios, model versions and various external artifacts such as spreadsheets, symbolic mathematics calculations, and external documentation. Failure to adequately manage such complexity can reduce the transparency, reliability, and credibility of the modeling process. While adherence to good modeling practices can aid this process, it often falls prey to corner-cutting or human error. This paper describes software that helps manage such complexity, by permitting modelers to easily access and succinctly compare historic versions of a model, by making explicit linkages between scenarios, the model versions and assumptions underlying them, and the motivations for and external files associated with model artifacts. 1 Introduction The System Dynamics modeling process offers great value in allowing stakeholders to rise above the welter of operational detail and perform long-term strategic planning. There is a certain irony in the fact that important higher-level insights can emerge from a modeling process that frequently generates and requires
    [Show full text]
  • Agent-Based Modeling and Simulation
    Proceedings of the 2009 Winter Simulation Conference M. D. Rossetti, R. R. Hill, B. Johansson, A. Dunkin and R. G. Ingalls, eds. AGENT-BASED MODELING AND SIMULATION Charles M. Macal Michael J. North Center for Complex Adaptive Systems Simulation Center for Complex Adaptive Systems Simulation (CAS2) 2 Decision & Information(CAS )Sciences Division Decision & Information Sciences Division Argonne National Laboratory Argonne National Laboratory Argonne, IL 60439 USA Argonne, IL 60439 USA ABSTRACT Agent-based modeling and simulation (ABMS) is a new approach to modeling systems comprised of autonomous, interacting agents. Computational advances have made possible a growing number of agent-based models across a variety of application domains. Applications range from modeling agent behavior in the stock market, supply chains, and consumer markets, to predicting the spread of epidemics, mitigating the threat of bio-warfare, and understanding the factors that may be responsible for the fall of ancient civilizations. Such progress suggests the potential of ABMS to have far-reaching effects on the way that businesses use computers to support decision-making and researchers use agent-based models as electronic laboratories. Some contend that ABMS “is a third way of doing science” and could augment traditional deductive and inductive reasoning as discovery methods. This brief tutorial introduces agent-based modeling by describing the foundations of ABMS, discuss- ing some illustrative applications, and addressing toolkits and methods for developing
    [Show full text]
  • Introduction Into Simulation Modeling for Business Applications
    AnyLogic 7 Dr. Andrei Borshchev, CEO George Meringov, Head of Development Nikolay Churkov, Technical Leader 1st AnyLogic Conference InterContinental Berlin December 2012 © The AnyLogic Company | www.anylogic.com Our ultimate goal: De-facto standard for simulation modeling (business applications) © The AnyLogic Company | www.anylogic.com 2 These things are NOT going to change: • Flexible general-purpose simulation software ̶ With extended support for some verticals • Java / Eclipse ̶ Cross-platform (Windows, Mac, Linux) ̶ Models on the Web ̶ Object-oriented (hierarchy, reuse of objects) • Multi-method, multi-language ̶ Discrete event, agent based, system dynamics ̶ Statecharts, process flowcharts, stock and flow diagrams, actioncharts… ̶ Discrete and continuous dynamics, events and differential equations • Logics linked to graphics but defined separately • Compatibility with v 6 fully preserved © The AnyLogic Company | www.anylogic.com 3 Key new features of AnyLogic 7 • “Merge and consolidation” • Uniform space markup for all kinds of activities • New library for discrete event (process) modeling August 2013 • New UI (focus on usability and minimum coding) • Extended support for "verticals“ • Scenario manager February 2014 © The AnyLogic Company | www.anylogic.com 4 Merge and consolidation: Objects Agent Agent3D Agent2D Entity Resource unit Active object © The AnyLogic Company | www.anylogic.com 5 Benefits for modelers • Use graphical editor to create entity internals: ̶ Fields ̶ Functions ̶ Animation • Define individual activities inside
    [Show full text]
  • Introducing Anylogic 7
    AnyLogic 7: New Features Overview Ilya Grigoryev Head of Training Services © The AnyLogic Company | www.anylogic.com AL7 New Features Overview. Agenda • AnyLogic UI & Usability • New Process Modeling library improvement ̶ “Pull” entity flow ̶ New workspace layout, more ̶ Flexible resource management / space for graphical editor alternative sets, preparation & wrap-up, ̶ Drop-down lists instead of typing shifts, breaks, preemption, preferences, … ̶ Probability distribution wizard, ̶ “Internals” of entities defined graphically custom distribution object • 3D • More support for agent based ̶ Gallery of more than 300 3D shapes modeling ̶ Painting 3D shapes ̶ True integration of modeling • Highly improved Pedestrian library methods (“The Great Merge”): ̶ Performance Entity = Resource unit = Agent ̶ Ease of use / markup ̶ Agent population wizard • Rail library improved ̶ Link object / network visualization ̶ Rail network construction / auto-connect, ̶ Inheritance between agent types curved segments, …. • Space markup ̶ Collection of US rail car 3D models ̶ Nodes, paths, walls, services, • Miscellaneous attractors, rail tracks, storages… ̶ Minor features ̶ One unified space for all objects ̶ AnyLogic roadmap © The AnyLogic Company | www.anylogic.com 2 Stays the same: • Powerful and flexible multi-method simulation environment ̶ Agent based / statecharts, decision rules, networks, … ̶ Discrete event / process flowcharts ̶ System Dynamics / stock and flow diagrams, numeric solvers • General-purpose “horizontal” tool with support for several verticals: ̶ Pedestrian ̶ Rail ̶ Health ̶ Logistics • Written in Java /cross-platform, open, exportable, embeddable ̶ Windows ̶ Linux ̶ Mac OS © The AnyLogic Company | www.anylogic.com 3 New version release. What are users afraid of? I won’t cope with converting my models to AL7 • All AL6 models run in AL7 • All AL6 libraries are still included • We have developed tools for migration AL 7 will override AL6 While migrating to AL 7, you can continue work in AL 6.
    [Show full text]
  • Evaluierung Der Simulationssoftware Anylogic, Arena Und SIMIO
    Evaluierung der Simulationssoftware AnyLogic, Arena und SIMIO Bachelorarbeit Abteilung Informatik Hochschule für Technik Rapperswil Frühjahrssemester 2013 Autor: Urs Baumann Betreuer: Prof. Dr.-Ing Andreas Rinkel Projektpartner: IBM Zürich Research Laboratory, Rüschlikon, ZH Experte: Dr. Ulrich Schimpel, IBM Zürich Research Gegenleser: Prof. Stefan F. Keller 1 ERKLÄRUNG Ich erkläre hiermit, dass ich die vorliegende Arbeit selber und ohne fremde Hilfe durchgeführt habe, ausser derjenigen, welche explizit in der Aufgabenstellung erwähnt ist oder mit dem Betreuer schriftlich vereinbart wurde, dass ich sämtliche verwendeten Quellen erwähnt und gemäss gängigen wissenschaftlichen Zitierregeln korrekt angegeben habe. das ich keine durch Copyright geschützten Materialien (z.B. Bilder) in dieser Arbeit in unerlaubter Weise genutzt habe. Ort, Datum: Name, Unterschrift: 2 AUFGABENSTELLUNG Ausgangslage, Problemstellung Im Modul System Modeling und Simulation wird momentan mit der Software Arena von Rockwell gearbeitet. Geplant ist, dass diese Software nächstes Jahr durch eine modernere Applikation, wie Simio oder AnyLogic abgelöst wird. Beide Programme verfolgen zum Teil unterschiedliche Modellierungsansätze und sind daher schwierig zu vergleichen. Im Rahmen der Bachelorarbeit sind die Unterschiede der beiden Softwarelösungen anhand eines universellen Beispiels aufzeigen und zu bewerten. Anschliessend sind die Ergebnisse mit den Erfahrungen von Arena gegenüber zu stellen. Ziele der Arbeit Inbetriebnahme der Software AnyLogic und Simio Einarbeitung
    [Show full text]
  • A Hybrid System Dynamics-Discrete Event Simulation Approach to Simulating the Manufacturing Enterprise
    A HYBRID SYSTEM DYNAMICS-DISCRETE EVENT SIMULATION APPROACH TO SIMULATING THE MANUFACTURING ENTERPRISE by MAGDY HELAL M.Sc. Benha Higher Institute of Technology, 1999 B.Sc. Benha Higher Institute of Technology, 1993 A dissertation submitted in partial fulfillment of the requirements for the degree of the Doctor of Philosophy in the Department of Industrial Engineering and Management Systems in the College of Engineering and Computer Science at the University of Central Florida Orlando, Florida Summer Term 2008 Major Professor: Luis Rabelo © 2008 Magdy Helal ii ABSTRACT With the advances in the information and computing technologies, the ways the manufacturing enterprise systems are being managed are changing. More integration and adoption of the system perspective push further towards a more flattened enterprise. This, in addition to the varying levels of aggregation and details and the presence of the continuous and discrete types of behavior, created serious challenges for the use of the existing simulation tools for simulating the modern manufacturing enterprise system. The commonly used discrete event simulation (DES) techniques face difficulties in modeling such integrated systems due to increased model complexity, the lack of data at the aggregate management levels, and the unsuitability of DES to model the financial sectors of the enterprise. System dynamics (SD) has been effective in providing the needs of top management levels but unsuccessful in offering the needed granularity at the detailed operational levels of the manufacturing system. On the other hand the existing hybrid continuous-discrete tools are based on certain assumptions that do not fit the requirements of the common decision making situations in the business systems.
    [Show full text]
  • Discrete Event Simulation on the Macintosh for Business Students - Agpss and Alternatives
    Proceedings of the 2016 Winter Simulation Conference T. M. K. Roeder, P. I. Frazier, R. Szechtman, E. Zhou, T. Huschka, and S. E. Chick, eds. DISCRETE EVENT SIMULATION ON THE MACINTOSH FOR BUSINESS STUDENTS - AGPSS AND ALTERNATIVES Ingolf Ståhl Department of Economic Statistics Stockholm School of Economics Box 6501 SE-Stockholm, SWEDEN ABSTRACT The paper first discusses the importance of discrete event simulation (DES) in the business school curriculum. It next notes how small Macintosh lap tops have become increasingly popular among business students. We next discuss what DES software is available on the Mac, first directly, then indirectly by running DES software for Windows in some way on the Mac. Noting that there is not much simple DES software on the Mac, but yet a great demand for such software from many business students, we turn to the transfer of one pedagogical software system, aGPSS, from Windows to the Mac. We here first give a brief historic background of aGPSS. Next we discuss some of the problems encountered when transferring aGPSS to the Mac. The paper ends with a brief discussion of some pedagogical aspects of using aGPSS on the Mac in the teaching of basic management science. 1 INTRODUCTION As a teacher of simulation at a business school, I have often been asked: Why do you teach discrete event simulation, DES, at a business school? Why are you not content by just simulating financial flows in a spreadsheet? The answer is that I have found DES, implying dynamic stochastic simulation, to be of great importance in a business curriculum for several reasons.
    [Show full text]
  • Getting Started with Anylogic and Agent Based Modeling Workshop Materials
    Getting Started with AnyLogic and Agent Based Modeling Workshop Materials Includes a step-by-step tutorial showing how to - import a Vensim system dynamics model into AnyLogic - integrate it with an agent based model - export the combined SD+AB model as Java applet International System Dynamics Conference Albuquerque, USA July 2009 © 2009 XJ www.anylogic.com Quick facts about AnyLogic • Sells since year 2000 • Current version: 6.4.1 – Editions: Advanced and Professional – Licenses: Commercial, Educational, University Researcher • The leading professional tool for AB modeling • The only multi-method tool supporting – System Dynamics – Agent Based Modeling – Discrete Event (Process-centric) Modeling – Any combination of these approaches within a single model • User base: 350+ companies, ~600 universities © 2009 XJ Technologies www.xjtek.com 2 Methods in simulation modeling Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System 1990s Agent Based © 2009 XJ Technologies www.xjtek.com 3 Methods in simulation modeling Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System System-level 1990s Agent Based Individual-centric © 2009 XJ Technologies www.xjtek.com 4 Methods in simulation modeling Process-centric System Dynamics (Discrete Event) Jay Forrester, 1950s Geoffrey Gordon, 1960s The System 1990s Agent Based Continuous, Discrete, Aggregated Disaggregated © 2009 XJ Technologies www.xjtek.com 5 AnyLogic OO Language Java/Eclipse platform Windows, Mac
    [Show full text]
  • What's So Great About Anylogic 6? a Bit of Anylogic Background
    What’s so great about AnyLogic 6? Dave Buxton ([email protected]) dseConsulting LTD UK agent for Sales, Project Support & Training ‘Happy to listen to all comments & questions and influence xjTEK’s product development’ A bit of AnyLogic background • xjTEK – Founded in 1992 & release AnyLogic 5 in 2000 – Version 6 launched in 2006 – xjTEK is 12 people • Plus N America • Plus Agents • User base is mainly… – Russia & North America – 140 known users – France & UK growing – 10 University customers • Plans within the UK – Book currently being written! – Web-based community forum (jointly with North American office) – Events • YOR, Warwick, 24 – 26 th March 09 • 1 day free workshop during spring – ‘getting started’ • ‘Official’ 3 or 5 day training course during Summer What’s it trying to achieve? • Multi-paradigm approach to simulation – It’s the only software that does this! • Models can contain approaches taken from Discrete Event & System Dynamics – Master of none? • Integrating Agent-based concepts is simple – Any body tried using swarm? – Handles complex systems • Change the focus from being on the modelling approach , start focussing on the problem What is AnyLogic? • An eclipse environment – handles time & the experiment – A series of Libraries that short-cut model development • The libraries are a way into AnyLogic, not the language itself! • Each block is actually just bits of AnyLogic code • Best illustrated by Version 5 • Model development is kept very close to the underlying Java – Advantages • Make simulation very powerful
    [Show full text]