Intelligent Agents = (Agent-oriented Design/Modeling/Simulation + Integration with AI)

Proposed Audience:

1) Anyone interested in Designing AI agents/applications, Integrating them into Simulation Models
2) Architects/Designers: Design AI agents and their interactions
3) Enterprise Architects/System Integrators: Design Systems of Intelligence, Automated/Intelligent Processes (RPA)
4) Consultants/Enterprise Architects: Digital Transformation of Business Processes/Organizations

Why is it important to learn how to build and use agent-based models (ABMs)? – Motivation, Objectives

An Agent is an autonomous individual element with properties and actions in a computer simulation.
Agent-Based Modeling (ABM) is the idea that the world can be modeled using agents, an environment, and a description of agent-agent and agent-environment interactions.

ABM is technically simple, but conceptually very deep. This unusual combination often leads to improper use of ABM.

If an ABM reproduces many characteristics of real-life scenarios, and its predictions are easy to understand, acceptance rates will be very high
//A Real-world Example applicable to “Managers” “Teams” “Healthcare” “Life Sciences” of day-to-day activities”

  • HR proposing employee-specific promotion, training recommendations
  • Team level delivery dynamics vs. Employee level resolution strategy
  • RWE for holistic treatment of an individual
  • PCOR replacing System level HEOR


What models are, and What modeling is – WHY do we build models anyway?

The modeling cycle – the iterative process of designing(formulating)/implementing/analyzing models and using them to solve scientific problems (Use mathematics and computer logic to rigorously explore the consequences of our simplifying assumptions)

WHAT agent-based models are – HOW are ABMs different from other knids of models – WHY would you use them?

Course Outline:


  • An Introduction to Agent-Based Modeling (ABM)
    • Provide a very brief introductor guide to carrying out a simple agent-based model (for Urban Health Research)
    • Formulate a problem, Formulate a model
      //Formulate a model that includes clusters of items and an individual “agent” that searches for the items in the model world
    • Sample Agents and Agent-based Models
    • Basic principles of agent-based simulation (learning, adaptation, emergence)
  • When to use ABM
  • When ABM is useful – Benefits of ABM
  • When to use ABM for real-world business problems (emergence, natural/flexible behaviour)
    • Four main application areas (flow simulation, organizational simulation, market simulation, diffusion simulation)
    • When is ABM actually useful?
      **ABM usage Checklist
    • Uses of ABM
  • ABM comparisons with other popular methodologies, terminologies
  • Agent-oriented Programming (AOP)
    • OOP and AOP Comparison, Overview of AOP framework
  • Limitations of ABM
  • Integrating Artificial Intelligence (AI) and Simulation Modeling
  • Issues in Agent oriented Software Engineering
    //Language Paradigms – Structured Programming, OOP, AOP (also Functional, Declarative)

//Urban Health Research (theory of approach + example of urban health)
//Case study of Medical Affairs in Pharma&MedTech

  • Relation between Complex Dynamic systems and Agents within it
  • How to Design an Agent-based-system (Model boundary, Subsystem diagram)
  • How to Design an Agent
    **Agent Design Checklist
    • Structure Assessment tests
  • How to Model Agent-based systems
    • Agent state/behavior/interactions (Agent structure, behavior, decisions, feedback loops, complexity)
      • Formulation of decision rules representing the behavior of agents
        • Principles for modeling decision making
      • Dynamic complexity (due to interactions among the agents over time)
      • When to use co-flows
    • Multi-Method Modeling
    • Embed AI into Simulation Models, and use in replacement of manually encoded behaviors or processes
  • How to convert an Object to an Agent
    • An Object-oriented Model to Agent-oriented Model
  • Pattern-oriented modeling (reference in Agent-based-modeling.pdf)
  • Noise in Systems Models/ABM
  • Common pitfalls in ABM
  • Gaps in Design & Implmentation (of Agents/Systems)


  • How to use AI techniques to address optimization/calibration problems in large scale agent-based models
  • How simulation models can integrate AI

— Tools —

  • AnyLogic
  • iThink (isee systems)
  • Eclipse STEM Modeler
  • NetLogo


  • Challenges for the future
  • Maturity Models for Systems Thinking