Now that you're comfortable using ExtendSim, let's get a little more in-depth into the advanced features in ExtendSim.
The following SimCasts, simulation experts and ExtendSim Simulation Engineers teamed up to divulge simulation techniques, discuss general research, and examine specific ExtendSim features. Through these webinars, you can learn tips and techniques that will help you achieve the maximum possible benefit from applying ExtendSim tools to your simulation projects.
Equation Block Basics
Equation-based block in ExtendSim allow you to enter equations that are simple logic statements (if x, then y) or complex calculations that control the model using ModL functions. In most cases, you won't need to debug the equations you write, but ExtendSim does provide an Equation Debugger to help you confirm your Equation blocks perform as expected.
This webinare is an introduction to ExtendSim Equation blocks. If you're not familiar with Equation blocks, it is highly recommended to view this video. It will acquaint you with the blocks and prepare you to watch the more advanced webinar "Equation Debugger".
Learn how to test equations you write in your equation-based blocks using the Equation Debugger.
- Set breakpoints to stop execution of an equation so you can step through the code.
- Set breakpoint conditions to cause the breakpoint to stop execution of an equation only under certain circumstances.
- Use the Step command to trace the execution of the equation and examine the effects on variables defined in the code.
Using the Reliability Module for Stand-Alone or Integrated RBDs
Dig into the Reliability module and come out with a sound understanding of its fundamental concepts, including:
- Basic Concepts and Definitions of Reliability
- Creating Distributions and Event Cycles
- Creating Reliability Block Diagrams
- Using RBD in Discrete Event or Discrete Rate Models
- Setting up Wear and Shifts in RBDs
Getting the Most Out of the ExtendSim Report Library
Meet the Report library introduced in ExtendSim 10! Learn how to set up and use each block in the Report library.
- Cost Stats • reports statistics for blocks that have costing enabled. Learn how to set up the costs in a model and how costs get reported.
- DB Statistics • calculates the statistics for a selected field in a table.
- Item Log Manager • collects information about how item states change over time for a customizable database report.
- Reports Manager • is the primary interface for creating reports of blocks and events.
- Statistics • reports statistics on certain types of blocks.
Transitioning to ExtendSim 10
Make a smooth transition to ExtendSim 10 by learning the primary differences between the two releases. In addition to watching this Transitioning to ExtendSim 10 video, check out the Upgrade Reference for ExtendSim 10 document. New features you'll learn about in these resources include:
- ExtendSim 10 product line & license types.
- Installing, launching, and activating ExtendSim 10.
- Dockable toolbars, tear off menus, and notebook windows.
- Adding and connecting blocks plus connection lines.
- Fastest vs multi-threaded run modes.
- Introduction to the Chart library.
- Introduction to the Report library.
- Announcing the Reliability module
- Transitioning existing ExtendSim models to ExtendSim 10.
- Transitioning custom ExtendSim blocks to ExtendSim 10.
- 64-bit app.
- Quick Start Guides, Tutorial & Reference guides, and videos.
- And more…
Managing Hierarchical Blocks
One of ExtendSim's most powerful and often used features is hierarchy. Learn about:
- Creating and modifying hierarchical blocks.
- Adding and removing connectors.
- Creating a user interface.
- Managing unique data.
Database Random Distributions
Optimization in ExtendSim
Optimization and simulation are two different tools used to solve different questions. However, there are times where the best approach would be to combine the two tools to solve different aspects of the same problem. ExtendSim gives a developer the power they need to use both in one tool.
Given that we are in the simulation community, we all know what discrete event simulation is. However many of us have misconceptions about what we mean when discussing optimization. This talk discusses in general terms some of the different optimization techniques and shows examples in ExtendSim.
It begins with a discussion about the Optimizer block in the ExtendSim Value library which uses a goal seeking approach. Learn how it works and what types of problems it works best on.
Then the discussion moves on to Linear Programming and Mixed Integer Programming. ExtendSim includes a Mixed Integer Linear Programming (MILP) solver and uses that technology in the Rate module. Using the development environment, we can also use the MILP solver in our own models during the simulation. After watching this, you will have a better understanding of optimization as well as when and how to incorporate it in your own models when the need arises.
The Ins and Outs of ExtendSim Messaging
ExtendSim uses a sophisticated messaging architecture to signal blocks into action. While messages can originate either from the ExtendSim application or from individual blocks, it is always a block that is on the receiving end of a message. Different types of messages result in the receiving block doing different types of things.
Introduction to Discrete Event Simulation
Edward J. Williams introduces discrete-event process simulation -- its concepts, usage, and importance in this SimCast. After watching this video, you will come to understand what discrete-event simulation is, how it can improve operations and fatten the bottom line, when to use it, and the fundamental steps of using it to maximum benefit. Additionally, you'll know what the next steps to take will be to begin enjoying the fruits of this powerful analytic technology.
Tools for Troubleshooting
Find the difference between what you want and what the model is actually doing. While this video discusses an earlier release of ExtendSim (ExtendSim 9), much of the techniques covered in the webinar are comparable in the current ExtendSim release. Block usage, the debugger, developing your own debugging methods, and other troubleshooting tools are all covered in this SimCast.
ConOps Modeling Using ExtendSim
Concept of Operations (ConOps) modeling methods and applications within ExtendSim are discussed in this SimCast. Topics covered include:
- An overview of the types of ConOps modeling • The objective in ConOps modeling is to measure the amount of equipment, spares, and/or fuel that might be needed to ensure a minimum amount of downtime. ConOps is also used to maximize the amount of processed material, packages, and/or searched area.
- The use of Excel to drive ExtendSim model behavior • The interconnect between ExtendSim and Excel provides the model architect with the ability to leverage the strengths of both tools. Learn how to solicit inputs and execute a model effectively even with vastly changing data sets.
- Methods and applications of the Query Equation blocks • Query Equation blocks offer a unique set of tools. Use these blocks and the ExtendSim database to control and change the state of the model.
- Model control and measurement • Queue Equation and Gate blocks can be used throughout a model to control items that represent different elements within the ConOps model. Adding blocks from the Rate library is a unique method of measuring continuous-rate resources such as fuel and power.
Discrete Event Simulation Using ExtendSim: Output Analysis
Problems encountered when analyzing simulation output that are modeled with queuing processes occur frequently. Two of the problems that typically occur are:
- The Initial Transient Period
- Autocorrelated Observations
Simulations take time to warm up to their steady state behavior, unless something is known about it when modeling a system. This warm up period or initial transient period must be accounted for in experimental design. Waiting times are not independent--they are autocorrelated--and these must also be dealt with in experimental design. It is easy to pick these problems apart by running a simple M/M/1 queue model. Learn more by watching this SimCast.
Other Tips for Your Simulation Projects
JMP Visualization Showcase
Learn how JMP® Exploratory Analytic Software can help bring out the story in your simulation results. Instead of relying only on spreadsheets for analysis, JMP saves you time by creating compelling visualizations that allow you to interact with the data and find the real drivers of performance. And it's simple to export data from ExtendSim. Just use the direct link to JMP from the ExtendSim Scenario Manager block.
Then it is easy to further analytically describe, compare, and predict with your simulated data using JMP's strong tool box, including design of experiments (DOE). This one-hour webinar features a real case study showing how JMP visualization can work in your simulation flow to save analysis time and take your results even further!
Satellite Orbiting and Communication Modeling
This SimCast looks at an orbital model, data communication model, and a detailed ground processing model. It examines polar orbiting weather satellites, stored sensor data, communication links to global receptors, and sensor data retrieval in ExtendSim models. The data retrieval model includes lost data and data recovery due to a rain attenuation model. The ground model is a process flow of sensor data executing simulated weather algorithms on shared data processors.
Soft Skills and Simulation: Lessons learned in using soft skills to avoid disaster and enhance your simulation project
"Ever since 1986 when I took my first graduate class in Discrete Event Simulation in SLAM (using Fortran IV as the underlying language), I have understood the tremendous power of simulation. I have been using simulation as one of my more dependable problem solving tools ever since. While many of my internal or external clients in the Army, Dept of Energy, Air Force, Pharmaceutical Companies, and Agriculture Companies have seen its power, some have not. I look back now over 37 years and blame any failures on:
- lack of communicating the process.
- focusing on my "cool programming".
- paying too little attention to simplicity vs realism tradeoffs.
- not having frequent enough inprogress demonstrations.
- inattention to the user's ability and eventual use of the model.
- not getting the company's objectives right.
The causes for failure were NOT bad mathematics, poor code, wrong methodology, or even wrong simulation package. This talk is about the soft factors, and the soft skills needed to ensure that your simulation project succeeds."
Inside Discrete-Event Simulation Software: How It Works and Why It Matters
Simulation practitioners and consumers can gain a grounding in how discrete-event simulation software works by viewing this SimCast. This Master Class follows Dr. Thomas J. Schriber and co-author Daniel T. Brunner's well-known paper "Inside Discrete-Event Simulation Software: How It Works and Why It Matters" which has been designated as one of ten Landmark Papers given in the first forty years of the Winter Simulation Conference. Topics include:
- Discrete-event systems
- Control elements and Operations
- Simulation runs
- Entity states
- Entity lists
- Entity list management
This SimCast concludes with several examples of "why it matters" to give modelers a better understanding of how their simulation software works. To get more out of this Master Class, we recommend you review Dr. Scriber's 2011 Winter Simulation Conference paper prior to viewing the SimCast. To get even further in-depth, Jerry Banks, Daniel T. Brunner, et al’s Handbook of Simulation (ISBN 0471134031) contains an extended version of this paper.
Getting to the Right Decision
Get set up for success as a Problem-Solver and help earn repeat business from your clients. This SimCast provides a framework for Modeling and Simulation projects that will show you the benefits of Front-End Contracting, good project management, and even when you shouldn't simulate.
Choosing the Best System
Experimenters are often interested in the problem of choosing the best from among a group of competing systems or alternatives. For instance:
- Which of three anti-cancer drugs has the highest cure rate?
- Which of five soft drink brands is the most preferred?
- Which of ten possible portfolios will yield the highest expected financial gain?
Such questions are also of interest to those working on simulation-based problems. For example:
- Which of three supply chain configurations will give our company the highest profit?
- Which of several production policies will maximize the probability that our orders will go out on time?
A number of ways to address problems such as these are discussed in this SimCast. Classical statistical methods are considered as well as more-modern ranking-and-selection techniques and even some stochastic optimization solution strategies. The talk is in the form of a survey. No sophisticated mathematical background is needed.
Design of Experiments for Simulation Modeling
Simulation models often have many input factors and determining which ones are really important can be quite difficult. This SimCast provides a brief introduction to design of experiments for simulation modeling.
Integrating ExtendSim with the Bayesian Network software package Netica
Dr. David P. Brown discusses how his company integrated ExtendSim with the Bayesian Network software package Netica (by Norsys Software). Integration of simulation with Bayesian belief network software offers several expanded capabilities for both ExtendSim builders and Bayes Net Builders. IDI used this on an Army project using ExtendSim to simulate a diesel generator to test a Bayesian network health monitoring system.
The Critical Importance of Simulation Input Modeling
An important part of any sound simulation study is that of modeling each source of system randomness by an appropriate probability distribution. This SimCast gives examples of data sets from real-world simulation studies, followed by a discussion of two critical pitfalls in simulation input modeling. The two major methods for modeling a source of randomness when corresponding data are available are delineated, namely, fitting a theoretical probability distribution to the data and the use of an empirical distribution. Finally, there is a discussion of how to model a source of system randomness when no data exists.
Quality Systems Modeling
Note: SimCasts on this page from 2018 and earlier were recorded prior to the release of ExtendSim 10. The concepts are the same and ExtendSim 10 can perform all these capabilities in these older videos plus more.