Model Testing and Verification

Course Overview

This session focuses on model verification — the process of ensuring that a simulation model is correctly programmed and behaves as intended. Participants will learn and apply both static and dynamic testing approaches using practical examples in AnyLogic.

---

Objectives

• Understand the key concepts of model verification
• Learn and apply static and dynamic testing approaches
• Use practical debugging techniques for simulation models in AnyLogic
• Identify common sources of errors in simulation models

---

Prerequisites

• Basic understanding of simulation modeling concepts
• AnyLogic: Discrete Events basics (Source, Queue, Delay, Sink), Functions, Events, System Dynamics (Stocks, Flows, Dimensions and Sub-Dimensions)
• Java syntax: variables, collections, basic data types, objects, conditions, error handling, loops, lists and maps
• Statistical distributions

---

Topics Covered

1. Introduction

• What is model verification and why it matters
• The distinction between verification and validation

2. Static Testing Techniques

• Analyzing code without executing it
• Structured walkthroughs and program correctness testing
• Examination of structural properties
• AnyLogic examples: module structure review and System Dynamics unit checks
• Activity: Analyze a pre-built AnyLogic model using static testing techniques

3. Dynamic Testing Techniques

• Executing the model under different conditions
• Traces and debugging
• Input-output relationship analysis
• Internal consistency checks
• Creating isolated test models for specific functions
• AnyLogic examples: varied input configurations, animation-based visualization, output data monitoring
• Activity: Execute and modify an AnyLogic model to investigate potential errors

4. Debugging Techniques

• Writing code in modules or functions for easier debugging
• Applying good programming practices and code reviews
• Tracking key variables, event times, and counters
• Comparing model results to known system behavior
• Using model animations for visual verification

5. Common Sources of Errors

• Data errors and conceptual model flaws
• Programming mistakes and implementation issues
• Random number generator problems

6. AnyLogic Debug Tool

• Using AnyLogic’s built-in debugging capabilities

---

Acquired Knowledge

By the end of this session, participants will be able to:

• Differentiate between static and dynamic testing methods
• Implement effective debugging techniques in AnyLogic
• Verify the correctness of simulation models by identifying and fixing errors systematically