Lakhasly

2.1 The learning perspective This tutorial teaches beginners to find their way through the Enterprise Dynamics (ED) simulation package.Later on, two servers are added in order to show the concept of "channels". The atoms used in this instance are Source, Queue, Server and Sink. Chapter 6 shows different methods to observe the results of a study. We use an example of a carpenter's factory, to explain the batching of products in Enterprise Dynamics. The user learns to use the: Monitor, the Summary Report, the new Experiment Wizard and displaying various graphics. Furthermore, the steps to go through in a simulation study are practiced at a beginner's level: model building, validation, experiment layout and analysis of results. In Chapter 7, the emphasis is placed on observing the pre-programmed methods through which products get access to a next atom (input strategy), are placed in a queue (queue discipline), or are redirected after use (send to). Chapter 8 introduces eight new atoms including the Assembler and the Conveyor atom, while Chapter 9 is dedicated to the link between Enterprise Dynamics and Excel. This chapter also provides a start in 4DScript and an introduction into labels. Chapters 9 is dedicated to the Human Resource atoms and Chapter 10 to Transporters. Besides this tutorial another important the tutorial is available: 'A first start in 4DScript'.This tutorial forms the starting point for learning the underlying programming language in Enterprise Dynamics. 2.4 Learning Simulation Simulation is not a simple technique. The sensible use of a simulation program requires knowledge about the theoretical background of discrete simulation such as probability distributions, the model building process including validation techniques and the design of experiments. Although the tutorial gives insights in this field, it is not a textbook regarding simulation in itself. For this, one can refer to existing books or lecture notes about discrete simulation. Developing a feeling for simulation and model building in particular are for the better part a matter of lots of practice. Examples of the models in this tutorial are also included in Enterprise Dynamics. Use them when in doubt whether your own models are correct. Enterprise Dynamics Tutorial 4 3 Enterprise Dynamics's Background Enterprise Dynamics is an object-oriented software program for modeling, simulation, visualization, and control of dynamic processes. The users can pick up objects - called atoms - from standard libraries in order to build their own model. Enterprise Dynamics is based on this concept of atoms as modeling objects in each model. An atom can represent a machine, a counter or a product but it can also have a non-physical character like a graph. As far as the difference in the type of atoms is concerned, we will mention basic atoms (five atoms often used: product, source, sink, server and queue), transport atoms (relating to transport such as conveyors or transporters), results atoms, etc. Thanks to the open structure of Enterprise Dynamics, the advanced user can build new atoms, for example to model a machine with very specific characteristics. At the moment, Enterprise Dynamics includes 100 standard atoms, but this number is ever increasing. The beginner may typically only need to select from around 30 frequently used atoms to have enough material for his or her applications. Atoms are predefined modeling objects used to build models quickly and to carry out studies. Enterprise Dynamics also has a built-in programming language called 4DScript, which can be used for setting specific conditions from reality in the model. In this tutorial we will address 4DScript only briefly. Thanks to the open structure of Enterprise Dynamics, the user can create its own library that should be loaded when Enterprise Dynamics is started using the packages structure of Enterprise Dynamics. It is also possible to create a new menu structure and layout. Enterprise Dynamics allows a problem solver to model virtually any problem and, by experimentation, look for a solution for a given problem or an answer to a specific question. Advantages of using Enterprise Dynamics within industry and logistics are: The ability to test a future system in an early design stage. Testing and improving proposed modifications resulting from e.g. Lean Manufacturing or Six Sigma studies, without impacting the operational environment. Modeling and analysis of several scenarios, such as Rough-Cut Capacity Planning, to be prepared for the future. Optimization and safeguarding of investment planning for production and transport equipment. Estimating the influence of uncertainties and variations, like failures and variable process times on system performances. Analysis and visualization of operational systems in 2D and 3D animation. Enterprise Dynamics Educational is our offering for educational purposes, consisting of a combination of the Enterprise Dynamics software with a number of elaborated case studies which the teacher can use for student assignments. This tutorial can be used to learn the basic skills necessary to work with Enterprise Dynamics. Furthermore it addresses topics and questions that typically arise in simulation studies and is therefore an important part of Enterprise Dynamics Educational. INCONTROL Simulations Solutions is a simulation solution provider: in addition to building, developing and selling the Enterprise Dynamics software, it also provides extensive services in the field of computer simulation, such as training and consulting. The consulting services involve carrying out simulation studies as well as application building. 4 Getting Familiar with Enterprise Dynamics 4.1 Starting Enterprise Dynamics Enterprise Dynamics can be started from the Windows Start Menu. During the startup a splash screen appears, see Picture 4-1. Enterprise Dynamics Tutorial 5 Picture 4-1: Splash screen When you start Enterprise Dynamics for the first time you are presented with the Example Wizard. The wizard offers a selection of the available example models that come with Enterprise Dynamics. Switch to the Examples page to view all the available models, see Picture 4-2. Later on, you can also open the Wizard from the Help menu. Just click on one of the models to open the example. Picture 4-2: Example page of the Example Wizard 4.2 The window sections As soon as Enterprise Dynamics is started up completely, the opening window should be approximately similar to Picture 4-3. Enterprise Dynamics Tutorial 6 Picture 4-3: Layout of the opening window in Enterprise Dynamics The window is divided into the following sections: A menu bar: among others for opening and saving files. Section 4.3 covers the menu structure more thoroughly. The library: The library includes all atoms a user can place into a model. Each atom has a certain function. By combining the right atoms, it is possible to re-create ('model') a business process in Enterprise Dynamics. Model building is described in chapter 5. The model layout window: this is where the model is being built. The run control: use this to reset and start the model and to regulate its execution speed. The clock displays the simulated time already elapsed during the simulation (not the real time!). 4.3 The menu structure The function and the appearance of the menus are similar to those in other Windows applications, such as Word and Excel. The most used menu options are explained in the table below. The main menu is to be found in the menu bar, which is subdivided as follows: File Make, open or save files, to set preferences and to control standard functions such as printing. Display Open viewer and layout windows to display models and to open the library windows with modeling objects. Simulate Open a Run Control or Clock window. Design and perform an experiment. Results To generate reports and graphics of a single simulation run or evaluate results of an experiment. Tools Contains tools such as the 4DScript interact and Autofit to fit a distribution to given data. Help Open the documentation, the Example Wizard as well as to find company and version information. Developer Tools useful for developer to create libraries and attribute functions. Each main menu item is divided into groups. See the documentation for a complete description. 4.4 Structure of the library and the model Enterprise Dynamics uses a tree structure to visualize the organization of atoms. With this structure the user can easily see which atoms contain which other atoms. The main tree gives a complete overview of the application, the library and the opened model. Two important trees are: Enterprise Dynamics Tutorial 7 The Library Tree (see Picture 4-4) which lists all atoms a user can insert into the model. The atoms are divided into groups, for example a transport group and a human resource group. By selecting an atom and dragging it into the model window (the 'Model Layout' or '2D Builder'), the atom is added to the model. The Model Tree, where all the atoms used in the current model are listed. Press F5 to refresh the list. Picture 4-4: The Library Tree & Model Tree of Example 1 The 2D Builder or Layout window and the Library Tree can be opened from the Display section of the main menu. The shortcuts are shown in Picture 4-5. Picture 4-5: Buttons to open the Library Tree and 2D Builder window. 5 Model Building Basics In the previous chapter, the theoretical aspects of Enterprise Dynamics have been discussed. In this chapter, we will start by building a simple model in Enterprise Dynamics. The objective here is to learn about Enterprise Dynamics, not to fully complete a simulation study. At the end of this chapter, the user should be able to develop a model in which several machines are used simultaneously. 5.1 Dragging atoms into the model The first step in the creation of a model is selecting the required atoms and placing these into the model. In this section, we will start by building a simple model that consists of the following four parts, see Picture 5-1: Source: The function of this atom is generating products into the model. Queue: This atom is a waiting area for customers or products. Server: The function of this atom is that of a machine or of a counter. Atoms entering a server undergo a process and remain in this atom for a certain time (the process time). Sink: The products or customers leave the model through this atom. Enterprise Dynamics Tutorial 8 Example 1 An average of 20 customers per hour come into a post office. The assistant has on average two minutes to help a customer. Of course, the number of customers may vary from one hour to another. The two-minute time that the assistant has to help a customer can vary as well. A customer who only wants to buy stamps will need less time than a customer who wants to open a new account. The customers are served on a first-in first-out basis. A few customers complained about the queues and because the post office manager is very concerned about the service to his customers, he wants to inquire into this problem. Questions and assignments

1. How high is the utilization of the post office assistant? What does that mean? 2. Can you make an estimate of the average queue length? 3. Which characteristics would you measure for the post office manager? Simulation is a tool to gain insight into the average queue length. Because this model is the first one we are going to build, it will be dealt with step by step. We therefore use the four atoms mentioned in the beginning of this section and put them into the model in the above-mentioned order. The first one is thus the Source, then the Queue, the Server and finally the Sink. These atoms and many others can be found in the Library. Just select an atom and drag it onto the Model Layout window. The four atoms that are required for this model can all be found in the first category "BASIC MODELING".In the Run Control window check the option "Run until stop time". Set the stop time on 8 and measurement on hours. If you now start the simulation it will stop after exactly 8 hours. Questions and assignments
2. Perform this a few times and note the number of windows produced. Does this correspond to the analysis carried out previously? What is the bottleneck in this process? 2. Results atoms The use of the results atoms is quite simple. The Library Tree (and not the Model Tree) has to be opened. Look for the StatusMonitor atom and the Generic Monitor atom in the Results group and drag them into the model. Note that the Generic Monitor can be found in the group Status within the Results group. To use the StatusMonitor, it is sufficient to connect its input channel to the central channel of the atom to be monitored. When the simulation is reset and started, the status monitor atom will automatically begin displaying the status statistics. In this case connect it to one of the bench vices, reset the model and press Start. Examine which proportion of its time the bench vice spends on waiting for timber to arrive, on collecting timber after the first arrived, and on actually processing the timber. If the channels are switched on, the atom should look like Picture 6-4. Enterprise Dynamics Tutorial 18 Picture 6-4: Part of the model with (already connected) Result atoms; left a StatusMonitor on the right a Generic Monitor. Configuring the Generic Monitor atom takes more work. Again, the first step is to connect the monitor with the atom you want to observe. In this case, we want to know how many windows are leaving the factory per hour. As a result, the monitor has to be connected with the Queue before the Sink. Connect the input channel of the monitor to the central channel (the information channel) of the Queue. You can also double-click or right-click the monitor atom and then press the button "Select Atom" select from the list that appears the atom that is going to be observed.7.2 Changing the Queue Discipline The purpose of this section is to change the Queue Discipline of the last queue and to observe theSelect "NegExp(10)". A 4DScript window will appear. Adjust the cycle time in such a way that the assistant needs an average of two minutes to help one customer. Picture 5-4: Input window Server Reset and restart the simulation with the Run Control. Because we have increased the time between two arrivals, it is a good idea to increase the simulation speed. By zooming in sufficiently, you can see how many customers are waiting in the queue. You can replace the blue dot, representing a customer in our example, with another icon. To do so, doubleclick the blue dot to the left of the source. Double-click on the edit field of the "2D Icon" that can be found on the "Visualization" tab.The model with this experiment is supplied under the name timber2.mod Enterprise Dynamics Tutorial 29 7 Playing with Strategies In this chapter, the emphasis lies on the pre-defined strategies with which products get access to a next atom (input strategy), are selected out of a queue (queue discipline) and are re-directed after processing (send to).It consists of three Sources, three Locks (an atom regulating the products' supply), five Servers, four Queues and one Switch, which is a self-built atom (Enterprise Dynamics offers this flexibility).Tutorials.2.3.2.2.3.4.1.3.1.2.3.4.5.6.

2.1 The learning perspective
This tutorial teaches beginners to find their way through the Enterprise Dynamics (ED) simulation package.
The tutorial starts in Chapter 4 with Enterprise Dynamics's opening window. From there and through a
progressive approach, you will get to know increasingly more about Enterprise Dynamics. This occurs mostly
through examples: small problems in a specific context. You will learn about the functionality of Enterprise
Dynamics by building and experimenting with models.
This approach has several advantages:

1. The user learns Enterprise Dynamics through modeling a simulation problem, which corresponds to the
   purpose for which the package is being used in practice.


2. The functionalities of a program are more effectively learned when they are connected to applications.


3. The user can refer back to several solutions in examples when he or she is confronted with a similar
   problem.
   Briefly, this tutorial aims not only to show what a function of the software does, but also how and when it can
   be applied. In addition, some parts of this tutorial can be used as a reference function, for example the
   overviews in the annexes.
   To be clear, this tutorial does not make the manual – available under the help menu (or when you press F1) –
   superfluous. Users who wish to reach a higher level in building simulation models will use the manual
   frequently, as it forms a complete documentation for all software functionalities.
   Enterprise Dynamics Tutorial 3
   2.2 Notation
   When a submenu is referred to, the symbol | is used. So File |Preferences refers to the submenu Preferences
   (consisting of a number of standard functions), which can be found under the File menu. Emphasis on
   important matters is rendered by bold or italic writing of words. The case studies and the related questions
   are written in this font. When 4DScript code is to be found in a text, this font has been used.
   A warning is preceded by: Warning! and an important tip, are displayed in a Tip box:
   Important tips are displayed in Tip boxes.
   2.3 Tutorial Structure
   Chapter 3 gives a short general introduction about the Enterprise Dynamics program. Chapter 4 explains
   Enterprise Dynamics's menu and window structure.
   The basic principles of the package are being demonstrated in chapter 5 by a simple queuing problem with just
   one server. Later on, two servers are added in order to show the concept of “channels”. The atoms used in this
   instance are Source, Queue, Server and Sink.
   Chapter 6 shows different methods to observe the results of a study. We use an example of a carpenter’s
   factory, to explain the batching of products in Enterprise Dynamics. The user learns to use the: Monitor, the
   Summary Report, the new Experiment Wizard and displaying various graphics. Furthermore, the steps to go
   through in a simulation study are practiced at a beginner’s level: model building, validation, experiment layout
   and analysis of results.
   In Chapter 7, the emphasis is placed on observing the pre-programmed methods through which products get
   access to a next atom (input strategy), are placed in a queue (queue discipline), or are redirected after use
   (send to).
   Chapter 8 introduces eight new atoms including the Assembler and the Conveyor atom, while Chapter 9 is
   dedicated to the link between Enterprise Dynamics and Excel. This chapter also provides a start in 4DScript
   and an introduction into labels.
   Chapters 9 is dedicated to the Human Resource atoms and Chapter 10 to Transporters.
   Besides this tutorial another important the tutorial is available: 'A first start in 4DScript'.This tutorial forms
   the starting point for learning the underlying programming language in Enterprise Dynamics.
   2.4 Learning Simulation
   Simulation is not a simple technique. The sensible use of a simulation program requires knowledge about the
   theoretical background of discrete simulation such as probability distributions, the model building process
   including validation techniques and the design of experiments. Although the tutorial gives insights in this field,
   it is not a textbook regarding simulation in itself. For this, one can refer to existing books or lecture notes
   about discrete simulation.
   Developing a feeling for simulation and model building in particular are for the better part a matter of lots of
   practice.
   Examples of the models in this tutorial are also included in Enterprise Dynamics. Use them when in doubt
   whether your own models are correct.
   Enterprise Dynamics Tutorial 4
   3 Enterprise Dynamics's Background
   Enterprise Dynamics is an object-oriented software program for modeling, simulation, visualization, and
   control of dynamic processes. The users can pick up objects – called atoms - from standard libraries in order to
   build their own model. Enterprise Dynamics is based on this concept of atoms as modeling objects in each
   model.
   An atom can represent a machine, a counter or a product but it can also have a non-physical character like a
   graph. As far as the difference in the type of atoms is concerned, we will mention basic atoms (five atoms
   often used: product, source, sink, server and queue), transport atoms (relating to transport such as conveyors
   or transporters), results atoms, etc.
   Thanks to the open structure of Enterprise Dynamics, the advanced user can build new atoms, for example to
   model a machine with very specific characteristics. At the moment, Enterprise Dynamics includes 100 standard
   atoms, but this number is ever increasing. The beginner may typically only need to select from around 30
   frequently used atoms to have enough material for his or her applications.
   Atoms are predefined modeling objects used to build models quickly and to carry out studies. Enterprise
   Dynamics also has a built-in programming language called 4DScript, which can be used for setting specific
   conditions from reality in the model. In this tutorial we will address 4DScript only briefly.
   Thanks to the open structure of Enterprise Dynamics, the user can create its own library that should be loaded
   when Enterprise Dynamics is started using the packages structure of Enterprise Dynamics. It is also possible to
   create a new menu structure and layout.
   Enterprise Dynamics allows a problem solver to model virtually any problem and, by experimentation, look for
   a solution for a given problem or an answer to a specific question. Advantages of using Enterprise Dynamics
   within industry and logistics are:
   The ability to test a future system in an early design stage.
   Testing and improving proposed modifications resulting from e.g. Lean Manufacturing or Six Sigma
   studies, without impacting the operational environment.
   Modeling and analysis of several scenarios, such as Rough-Cut Capacity Planning, to be prepared for
   the future.
   Optimization and safeguarding of investment planning for production and transport equipment.
   Estimating the influence of uncertainties and variations, like failures and variable process times on
   system performances.
   Analysis and visualization of operational systems in 2D and 3D animation.
   Enterprise Dynamics Educational is our offering for educational purposes, consisting of a combination of the
   Enterprise Dynamics software with a number of elaborated case studies which the teacher can use for student
   assignments.
   This tutorial can be used to learn the basic skills necessary to work with Enterprise Dynamics. Furthermore it
   addresses topics and questions that typically arise in simulation studies and is therefore an important part of
   Enterprise Dynamics Educational.
   INCONTROL Simulations Solutions is a simulation solution provider: in addition to building, developing and
   selling the Enterprise Dynamics software, it also provides extensive services in the field of computer
   simulation, such as training and consulting. The consulting services involve carrying out simulation studies as
   well as application building.
   4 Getting Familiar with Enterprise Dynamics
   4.1 Starting Enterprise Dynamics
   Enterprise Dynamics can be started from the Windows Start Menu. During the startup a splash screen appears,
   see Picture 4-1.
   Enterprise Dynamics Tutorial 5
   Picture 4-1: Splash screen
   When you start Enterprise Dynamics for the first time you are presented with the Example Wizard. The wizard
   offers a selection of the available example models that come with Enterprise Dynamics. Switch to the
   Examples page to view all the available models, see Picture 4-2. Later on, you can also open the Wizard from
   the Help menu. Just click on one of the models to open the example.
   Picture 4-2: Example page of the Example Wizard
   4.2 The window sections
   As soon as Enterprise Dynamics is started up completely, the opening window should be approximately similar
   to Picture 4-3.
   Enterprise Dynamics Tutorial 6
   Picture 4-3: Layout of the opening window in Enterprise Dynamics
   The window is divided into the following sections:
   A menu bar: among others for opening and saving files. Section 4.3 covers the menu structure more
   thoroughly.
   The library: The library includes all atoms a user can place into a model. Each atom has a certain
   function. By combining the right atoms, it is possible to re-create (‘model’) a business process in
   Enterprise Dynamics. Model building is described in chapter 5.
   The model layout window: this is where the model is being built.
   The run control: use this to reset and start the model and to regulate its execution speed.
   The clock displays the simulated time already elapsed during the simulation (not the real time!).
   4.3 The menu structure
   The function and the appearance of the menus are similar to those in other Windows applications, such as
   Word and Excel. The most used menu options are explained in the table below.
   The main menu is to be found in the menu bar, which is subdivided as follows:
   File Make, open or save files, to set preferences and to control standard functions such as printing.
   Display Open viewer and layout windows to display models and to open the library windows with modeling
   objects.
   Simulate Open a Run Control or Clock window. Design and perform an experiment.
   Results To generate reports and graphics of a single simulation run or evaluate results of an
   experiment.
   Tools Contains tools such as the 4DScript interact and Autofit to fit a distribution to given data.
   Help Open the documentation, the Example Wizard as well as to find company and version
   information.
   Developer Tools useful for developer to create libraries and attribute functions.
   Each main menu item is divided into groups. See the documentation for a complete description.
   4.4 Structure of the library and the model
   Enterprise Dynamics uses a tree structure to visualize the organization of atoms. With this structure the user
   can easily see which atoms contain which other atoms. The main tree gives a complete overview of the
   application, the library and the opened model.
   Two important trees are:
   Enterprise Dynamics Tutorial 7
   The Library Tree (see Picture 4-4) which lists all atoms a user can insert into the model. The atoms are
   divided into groups, for example a transport group and a human resource group. By selecting an atom
   and dragging it into the model window (the ‘Model Layout’ or ‘2D Builder’), the atom is added to the
   model.
   The Model Tree,
   where all the atoms used in the current model are listed.
   Press F5 to refresh the list.
   Picture 4-4: The Library Tree & Model Tree of Example 1
   The 2D Builder or Layout window and the Library Tree can be opened from the Display section of the main
   menu. The shortcuts are shown in Picture 4-5.
   Picture 4-5: Buttons to open the Library Tree and 2D Builder window.
   5 Model Building Basics
   In the previous chapter, the theoretical aspects of Enterprise Dynamics have been discussed. In this chapter,
   we will start by building a simple model in Enterprise Dynamics. The objective here is to learn about
   Enterprise Dynamics, not to fully complete a simulation study. At the end of this chapter, the user should be
   able to develop a model in which several machines are used simultaneously.
   5.1 Dragging atoms into the model
   The first step in the creation of a model is selecting the required atoms and placing these into the model. In
   this section, we will start by building a simple model that consists of the following four parts, see Picture 5‑1:
   Source: The function of this atom is generating products into the model.
   Queue: This atom is a waiting area for customers or products.
   Server: The function of this atom is that of a machine or of a counter. Atoms entering a server undergo a
   process and remain in this atom for a certain time (the process time).
   Sink: The products or customers leave the model through this atom.
   Enterprise Dynamics Tutorial 8
   Example 1
   An average of 20 customers per hour come into a post office. The assistant has on average two minutes to help
   a customer. Of course, the number of customers may vary from one hour to another.
   The two-minute time that the assistant has to help a customer can vary as well. A customer who only wants to
   buy stamps will need less time than a customer who wants to open a new account. The customers are served
   on a first-in first-out basis.
   A few customers complained about the queues and because the post office manager is very concerned about
   the service to his customers, he wants to inquire into this problem.
   Questions and assignments


4. How high is the utilization of the post office assistant? What does that mean?


5. Can you make an estimate of the average queue length?


6. Which characteristics would you measure for the post office manager?
   Simulation is a tool to gain insight into the average queue length. Because this model is the first one we are
   going to build, it will be dealt with step by step.
   We therefore use the four atoms mentioned in the beginning of this section and put them into the model in the
   above-mentioned order. The first one is thus the Source, then the Queue, the Server and finally the Sink.
   These atoms and many others can be found in the Library. Just select an atom and drag it onto the Model
   Layout window. The four atoms that are required for this model can all be found in the first category “BASIC
   MODELING”.
   Click the “+” belonging to the first category “BASIC MODELING” to open the section, then drag a Source, then
   the Queue, the Server and finally the Sink in the model. When this is done, you get to see Picture 5‑1 on your
   screen:
   Picture 5-1: The first model
   In case of errors, click on the atom concerned and press ‘delete’ to remove the atom from the model.
   To check whether the customers walk through the model in the correct sequence, we will start a simulation
   run using the Run Control window. If this window is not visible yet, switch to the Simulate page in the main
   menu and press on the Run Control button. In the Run Control window, the sub-function Slide Control has to be
   selected, see Picture 5‑2. This function enables you to adjust the speed of the simulation.
   Picture 5-2: Select Slide Control
   Before the simulation can start, you first have to click on the Reset button. It is the leftmost button in the Run
   Control window. After clicking on the Reset button, a blue dot appears next to the Source atom. This is a
   product atom. By clicking on the start button (green triangle) the products, in this case customers will pass
   Enterprise Dynamics Tutorial 9
   through the other atoms, in this case the post office. If necessary, adjust the speed by dragging the speed
   button in the Run Control window!
   By right-clicking anywhere in the Run Control window a context menu will pop up with the following options:
   Stay On Top. If this option is checked, the Run Control window will always be on top of other
   windows, including windows of other applications.
   Show Settings. With this option you can hide and show the settings tabs in the Run Control window.
   In this model no queue will form before the counter (Server). As a result, the blue dot is only visible in the
   counter (Server). The percentage displayed in the Server reflects the utilization of the Server so far. When the
   percentage is not visible, you might need to zoom in or zoom out. Click on the left and right mouse buttons
   simultaneously and move the mouse forwards and backwards to zoom in or out.
   The atoms have been placed in the right order, but we still need to enter how many customers per hour are
   coming in and how long the assistant needs to help a customer. Before you enter these cycle times in
   Enterprise Dynamics, you have to know that all cycle times in Enterprise Dynamics are defined in seconds.
   When the post office assistant needs two minutes to help a customer, you enter 120 seconds in Enterprise
   Dynamics. Equally, we must define how much time elapses between customer arrivals, called the Inter-arrival
   time, in seconds.
   Picture 5-3: Input window Source
   First, we change the parameters of the Source, so that an average of 20 new customers per hour arrive at the
   post office. Right-click (or double-click) the Source atom, an input window appears, see Picture 5‑3. On the
   “General” tab the most important parameters of the Source can be found. The time elapsing between two
   arrivals can be entered in the Inter-arrival time field. We use a probability distribution to model the
   uncertainty in the arrival process. In this example, the negative exponential distribution is used, which is a
   probability distribution often applied in unpredictable arrival or service processes. In Enterprise Dynamics, the
   following 4DScript code has to be entered: NegExp(e1), where the parameter e1 stands for the average value
   or expectation of the exponential distribution. Now enter 180 as inter-arrival time and confirm with OK.
   Important: To the right of the input fields, you will often find a triangle pointing downwards. Clicking or
   doubleclicking on this triangle opens a list with available and pre-defined options. First check whether any of
   these options might be useful for your specific case! For an explanation of these options see the Enterprise
   Enterprise Dynamics Tutorial 10
   Dynamics Help file. The Help contains a full description of the most important atoms and their input fields.
   The grey letters ‘4DS’ indicate that 4DScript commands, Enterprise Dynamics’s programming language, can be
   entered.
   The time that an assistant needs to help one customer has to be entered as well. Again, the negative
   exponential distribution is used. Now, right-click or double-click the Server atom, so that the service time can
   be adjusted on the “General” tab sheet, see Picture 5‑4. Select “NegExp(10)”. A 4DScript window will
   appear. Adjust the cycle time in such a way that the assistant needs an average of two minutes to help one
   customer.
   Picture 5-4: Input window Server
   Reset and restart the simulation with the Run Control. Because we have increased the time between two
   arrivals, it is a good idea to increase the simulation speed. By zooming in sufficiently, you can see how many
   customers are waiting in the queue.
   You can replace the blue dot, representing a customer in our example, with another icon. To do so, doubleclick the blue dot to the left of the source. Double-click on the edit field of the "2D Icon" that can be found on
   the “Visualization” tab. The Resource Manager will appear, featuring (some of) the icons the users can select.
   Click on the icon you want to use and select OK: the blue dot in the Product input window is replaced by your
   new icon. Confirm this change with OK.
   The icon that Enterprise Dynamics uses in the 3D display can be changed as well. Use for this the option 3D
   Icon and select the Person icon from the list. Now display the model in 2D and 3D. To open the 3D
   visualization: select the Display page of the main menu and click on the 3D Viewer or 3D Builder button. To
   navigate through the 3D model, move the mouse and click on the mouse buttons.
   Enterprise Dynamics Tutorial 11
   There are many more icons available with the Enterprise Dynamics installation. You can also add your
   own icon files to the Resource Manager and use these in your model. Click on the Resource Manager
   button on the Tools page of the main menu to open this window. Click on Add, a window appears.
   Browse and select an icon and click OK to add it to the Resource Manager. You then can use this icon as
   the 2D icon of a product.
   In this case, the icon is only changed to improve the visualization. However, if distinguishing between different
   types of customers is necessary, the use of several icons can bring additional clarity. For example, one group
   of customers could be represented by a green icon and another by a blue icon.
   The model can now be saved, with File|Save as. For reference purposes, the model of this post office can be
   found in the folder tutorial models under the name Postoffice1.mod.
   5.2 The Channels
   Example 1 (continued)
   In the neighborhood, two post offices have been shut down. The manager hopes that many new customers are
   going to use his post office and even expects twice as many customers in the near future. As a result, an
   average of 40 customers will come into his post office per hour. In order to figure out if his post office can
   manage this flow with the existing counter, he decides to carry out a new simulation study. Questions and
   assignments


7. Does it make any sense to simulate this situation? In order to answer this question, use the expected
   utilization and check your own supposition by modifying the model postoffice1.mod to the new situation
   and by carrying out a simulation run.
   Example 1 (continued)
   It seems that one counter will not be enough to manage the flow. The manager now wants to open 3 counters
   to be on the safe side. Once again, he wishes to see the effects of this adjustment on the queue.
   To begin with, we put the two counters under the existing counter (select the right atom to model a counter).
   Then, the parameters for the counters’ cycle times have to be changed in such a way that they correspond to
   the first counter. To this end, you might want to go back to chapter 5.1. The model should now look like
   Picture 5‑5.
   Picture 5-5: Post office with three counters
   Atoms can simply be copied –including all entered fields and parameters-! For this, first click on the atom
   to be copied and then on F6: a duplicate of your atom is created. You can also use Ctrl+C to copy and
   Ctrl+V to paste the atom.
   To copy more than one atom right-click and select an area around these atoms or hold down the Ctrl-key
   and click on the atoms to add these to your selection: click F6 or use Ctrl+C to copy the selection and
   Ctrl-V to paste it. You can reposition the entire selection at once.
   When the simulation is started, you will see that the customers are still sent to the first counter. In order to
   correct this, we have to look into the method that Enterprise Dynamics uses to direct the customers.
   In the View menu of the Model Layout window, select the option Channels, see Picture 5‑6. By doing so, the
   channels of the atoms appear. The purpose of these channels in Enterprise Dynamics is to route products and
   to transmit information.
   Enterprise Dynamics Tutorial 12
   Picture 5-6: Switching on Channels
   Switch on the option enabled to see the channels, see the result in Picture 5‑7.
   Picture 5-7: The original model of the post office with channels switched on
   When the lines between two channels do not appear curved but squared, it might be helpful to switch on the
   option Channels as Arches in the View | Channels, see again Picture 5‑6. As soon as the channels are switched
   on, a yellow box is shown in the upper left corner of all atoms. This box contains the input and output
   channels of the concerned atom, see Picture 5‑8.
   Picture 5-8: Connected channels
   The box contains several dots, the left dot is an input channel and the right one an output channel. A channel
   can be open (in green) or closed (in red). When both the input and output channels are open, the connection
   between the input and output channels is ready (in green), otherwise, the connection is not ready (in red).
   Products come into an atom through an input channel and leave the atom through an output channel. An input
   channel has to be connected to an output, whereas an output channel has to be connected to an input. Each
   input or output channel can only be connected to one other channel.
   By clicking on the “+” sign next to a dot, the number of input or output channels can be increased. By
   rightclicking a channel dot, an overview of all connections of the atom appears. The “-” sign (which appears as
   soon as the number of input and output channels is two or more) speaks for itself.
   The yellow dot in the middle of the channel box is designed to receive (but not send!) information, this
   channel is called the Central Channel. The central channel cannot be used to send product atoms to other
   atoms but used to register information regarding an atom. Later we will see that the input channel of a result
   atom can be connected to the central channel. Both input channel and output channels can be connected to
   central channels. There is always only one central channel present on an atom, but several atoms can be
   connected to the same central channel.
   Enterprise Dynamics Tutorial 13
   Important: Although you can use channels for many purposes, the basic function of channels is the indication
   on an atom (read: queue, counter) of all possible next locations for products in this atom. In this way, the
   channels provide a (rough) routing of products through the model!
   Picture 5-9 represents an atom with 4 input channels, 6 output channels and the central channel.
   Picture 5-9: Atom with 4 input channels and 6 output channels
   One of the revisers of this document happens to be color-blind, (as are most laser printers) and therefore
   has trouble distinguishing an open (green) and closed (red) channel. You can change these color settings
   in Enterprise Dynamics yourself by clicking File|Preferences …|Appearances. Change the color of “2D ->
   Open channel” and “2D -> Closed channel”.
   By dragging a line between an output and an input channel with the mouse cursor, two atoms are connected
   with each other. By dragging a line between an input or an output channel and the central channel of the same
   atom, a connection is broken.
   Now connect the channels in the Enterprise Dynamics model in such a way that the customers only go to one of
   the three counters and afterwards from one counter to the exit. The result should look like Picture 5‑10.
   Restart the simulation (do not forget to reset!) and check if the customers now go to the right counters.
   Picture 5-10: Post office with connected channels
   The reason why the customers are still only going to the first counter is that unless we tell Enterprise Dynamics
   otherwise, it sends all products through output channel 1. In the input window that appears by double-clicking
   or right-clicking an atom, you can specify the channel through which the products should leave the atom. By
   doubleclicking on the Queue, we can alter the parameters of our model. In this case, we have to adjust the
   value in the Send To field, see Picture 5-11: if we want that each open channel can be selected, the
   predefined options 2, 3 and 4 are relevant.
   Post office with connected channels
   Enterprise Dynamics Tutorial 14
   Picture 5-11: The list of predefined option for the "Send To" property of the Queue
   Explain why in option 2 or option 3, either the first or last server will get the highest utilization, while the
   servers’ utilizations in option 4 will be more or less the same after the simulation has run for some time!
   In the Send To field, you can either enter a number, write a short piece of 4DScript code that will result in the
   output channel number, or use predefined code. For beginners, the last option is definitely the easiest
   method. By clicking on the small down triangle in the input field, a list appears, showing the predefined
   options that the user can select. After a certain option is selected, text displayed in blue can be modified.
   Obviously, this possibility is only applicable if there is something left to change.
   Now select a strategy in which the utilization of the 3 counters is about the same. If the utilizations of the
   counters are not directly visible, use both mouse buttons in order to zoom in or zoom out. View the model also
   in 3D with Display | 3D Viewer or Display | 3D Builder!
   Picture 5-12: The 3D-model with the three counters modeled as machines
   Questions and assignments


8. Is the layout with 3 counters sufficient?
   The model can be saved now. For reference purposes, the model of this post office is to be found in the folder
   tutorial models under the name Postoffice2.mod.
   Enterprise Dynamics Tutorial 15
   6 Analyzing the Results
   In the last chapter we used a detailed and progressive method to simulate a post office. You learned how to
   place atoms in a model, and how to connect the channels between atoms. You also learned how to create a
   flow of product atoms into the model at a defined interval, and how to make them stay on a machine
   (counter) for the correct processing time. As a result, all the basic principles for building a model have been
   dealt with.
   In this chapter, we are going to build a more complex model and above all observe how to follow and measure
   the results of a simulation study. We will once again start with a (simple) example and (after building the
   model) explain the different ways of measuring results. Next, we will practice this in the model. Example 2:
   The carpenter’s factory General context:
   A carpenter’s factory often has its own design department. This department is complex to manage because
   each order consists of a separate project, in which the process times of each production step has to be
   estimated as accurately as possible. Large intermediate stocks, long throughput times, and variable
   bottlenecks often characterize such an environment.
   The management of a carpenter’s factory, which mainly produces windows and window frames, wishes to gain
   insight into bottlenecks, production numbers and throughput times. For this purpose, they decide to have a
   simulation study carried out. To keep it simple, in this case only the production of windows is included.
   The window production process can be divided into several steps. First, long timbers come in, which are cut
   into 10 shorter lengths on a saw. Next, these shorter pieces are cut into the desired pattern on two milling
   machines. When the parts have been milled, they go to the bench vice.
   At this stage, four pieces at a time are placed in a frame and glued. There are two milling machines and two
   bench vices, operating in parallel, so the parts pass through milling and gluing only once. There are storage
   areas between all consecutive production steps, and there is a continuous supply of raw material. To control
   the stock levels, each intermediate storage space is limited to 100 parts.
   The time to saw the incoming timber into shorter lengths is uniformly distributed between two and three
   minutes and the processing time on the milling machine is normally distributed with an average of 36 seconds
   and a standard deviation of 2 seconds.
   The bench vice takes exactly two minutes to glue a window together. The production process takes place
   continuously between 09:00 to 17:00 hours. The products that are not completed at the end of the day remain
   until the next day.
   Questions and assignments: Analysis beforehand


9. Make a drawing of the process with the capacity per hour in raw lengths, milled lengths and frames for
   each production stage.


10. How many window frames are expected to leave the factory each day?
    Now build the model in Enterprise Dynamics. The probability distributions can be found in the drop down menu
    of the cycle time. Note that for the normal distribution, the formula max(0, normal(36,2)) is used to
    prevent negative service times. Change the Batch rule on the “Specific” tab of the Servers; use the Batch Rule
    “1 in, B out” with batch size 10 on the servers for the 1:10 operation from timbers to short lengths and the
    batch rule “B in, 1 out” with batch size 4 for the 4:1 operation from short lengths to frames. Change the name
    of the Servers, so that the model is easier to read. If everything went well, your model should look like the
    model in Picture 6-1.
    Picture 6-1: Layout of the carpenter’s factory
    Changing the size of the atoms can be done by clicking the atom, and then dragging its upper left or the lower
    right corner.
    Enterprise Dynamics Tutorial 16
    The Product atom (the blue dot) can be changed into a brown `timber’: double-click or right-click on the
    product and change the input field as indicated below.
    Picture 6-2: Changing the product’s general settings
    Picture 6-3: Changing the product’s display
    Save the changes you have made. Which changes did you make?
    The 2D icon with the blue dot is still there, but if 'Show 2D icon' is unselected only the brown background
    remains visible. The product has a length of 0.2 meters and a width of 1 meter, similar to a timber in upright
    position. In 3D, the height (size z) of 0.2 meters plays a role. Note that the product icons can differ in 2D and
    3D. In 3D, select e.g. the pallet or another icon!
    In 2D, the grid visible behind the atoms functions as a grid for coordinates. Each square represents an area
    with a size of 1 x 1 meters. The black square marks the coordinates (0,0). Physical dimensions do not play a
    role in this abstract model of a carpenter's factory, but they definitely do play a role in other applications such
    as warehouses with short walking distances, or fork lifts having to cover a certain distance!
    If you are not sure that you got your own model correct, you can also use the supplied model timber1.mod in
    the next sections.
    6.1 Techniques to measuring the results
    Now that we have a working model, we can start with the measuring results. Within Enterprise Dynamics,
    there are several techniques that can be used:
    Enterprise Dynamics Tutorial 17


11. Information indicated on the atoms.
    Each Queue shows how many products are in the queue, each Server shows its utilization and each
    Source or Sink shows how many products have entered or left the model. This information is particularly
    useful to see if a model is working logically during the simulation (which is part of the verification of the
    model).


12. The Result Atoms.
    The Results atoms can be found in the Results section of the library. For example, the StatusIndicator
    and StatusMonitor atoms display the current status of an atom and the fraction of time that the atom is
    Busy, Idle or Blocked.
    The Generic Monitor Atom can graphically display other information about an atom. The user can select
    which information is to be shown, and in which form. This atom is particularly useful to validate the
    model.


13. The Summary Reports and the Graphs from the Results menu.
    The intermediate results of a simulation run are shown here. This method is especially convenient for a
    quick overview of the system’s status and a rapid feedback of the effect of changing various
    parameters.


14. Setting up an experiment with the Experiment Wizard.
    This technique is quite different from the last three and is used for the actual study. The length of a
    measurement period and how often this period is to be simulated, e.g. 10 times half a year, is set
    beforehand. We must also specify what variables we want to measure. At the end of the experiment the
    mean values of the variables are given within confidence limits.
    6.2 Measuring the results
    In section 6.1, four techniques have been presented, through which the results of a simulation study can be
    shown. In this section, we are going to apply these four methods to the carpenter’s factory. For this, the
    carpenter’s factory model has to be opened in Enterprise Dynamics.


15. Information shown on the atom
    The clock function in Enterprise Dynamics has to be visible on the screen (submenu of Simulate). Start the
    simulation and see how many products leave the factory within eight hours. In the Run Control window check
    the option “Run until stop time”. Set the stop time on 8 and measurement on hours. If you now start the
    simulation it will stop after exactly 8 hours.
    Questions and assignments


16. Perform this a few times and note the number of windows produced. Does this correspond to the
    analysis carried out previously? What is the bottleneck in this process?


17. Results atoms
    The use of the results atoms is quite simple. The Library Tree (and not the Model Tree) has to be opened. Look
    for the StatusMonitor atom and the Generic Monitor atom in the Results group and drag them into the model.
    Note that the Generic Monitor can be found in the group Status within the Results group.
    To use the StatusMonitor, it is sufficient to connect its input channel to the central channel of the atom to be
    monitored. When the simulation is reset and started, the status monitor atom will automatically begin
    displaying the status statistics. In this case connect it to one of the bench vices, reset the model and press
    Start. Examine which proportion of its time the bench vice spends on waiting for timber to arrive, on collecting
    timber after the first arrived, and on actually processing the timber. If the channels are switched on, the atom
    should look like Picture 6‑4.
    Enterprise Dynamics Tutorial 18
    Picture 6-4: Part of the model with (already connected) Result atoms; left a StatusMonitor on the right a
    Generic Monitor.
    Configuring the Generic Monitor atom takes more work. Again, the first step is to connect the monitor with the
    atom you want to observe. In this case, we want to know how many windows are leaving the factory per hour.
    As a result, the monitor has to be connected with the Queue before the Sink. Connect the input channel of the
    monitor to the central channel (the information channel) of the Queue. You can also double-click or right-click
    the monitor atom and then press the button “Select Atom” select from the list that appears the atom that is
    going to be observed.
    Secondly, you have to enter what is going to be observed. For this, double-click or right-click once again on the
    monitor atom and adjust the “Variable to Monitor”. Select the option “Output per hour” from the list of
    predefined options.
    If the requested variable does not appear in the list, 4DScript code has to be entered. To this end, select the
    option 4DScript Expression from the list and click the square button to open a 4DScript Editor. You can now
    enter your own code. However, this is not recommended for beginners.
    Now you see why we added the Queue before the Sink: to use the pre-defined statement Output per hour (a
    Sink has no output!).
    Enterprise Dynamics Tutorial 19
    Picture 6-5: Monitor variable
    Set the Sample rate to 60 seconds: measurements will now be taken every (simulated) minute. Because the
    average output is not likely to be restricted to whole numbers change the number of decimals to 1. Try Auto
    adjust, Maximum value, to develop a feeling for the various settings!
    We choose 100 as maximum value as you can see in Picture 6‑5. Now start the simulation.
    Using a large number of monitor atoms with the sample rate set to the default value of 5 seconds can be
    the cause of models running very slowly! Remember that the monitor atom is being displayed every 5
    simulated seconds, and thus is taking up a lot of processing time! This phenomenon even occurs when
    the model layout window is closed. In many instances, a sample rate of 60 seconds or even 3600 seconds
    is preferable.
    Therefore use the default sample rate of 5 seconds in monitor atoms with caution.
    Questions and assignments


18. How many windows are leaving the factory on average per hour?


19. Summary report and graphs
    A third method to measure results is to use the option 'Summary report' and 'Graphs' from the Results menu.
    Enterprise Dynamics Tutorial 20
    We will first deal with the option Summary reports. It enables you to get an overview of the status of the
    model, where a distinction is made between the following features:
    Current Content: The number of products present in an atom the moment the report is made.
    Average Content: The average number of products that were present in the atom.
    Throughput Input: The number of products that have entered the atom.
    Throughput Output: The products or customers leave the model through this atom.
    Staytime average: The average time the products are spending in the atom.
    By selecting the option Summary Report in the Results menu, a report such as in Picture 6‑6 appears.
    Picture 6-6: Summary report
    Accordingly, you can see in Picture 6‑6 that a product is present on average 36 seconds in milling-machine 1,
    that 992 products have been produced in bench vice 2 and that 4 products are still present there.
    As you can see in the lines of the saw and the two bench vices, all measurements on servers with batch
    transitions are performed in the “smaller” unit, i.e. the units each batch consists of. For instance, look at the
    output of Source1 of 210, and the input of the Saw of 2100. The measurements on the saw therefore apply to
    short timbers rather than long timbers, and the measurements on the bench vices apply to the milled short
    timbers rather than window frames.
    It would have been possible that the output (not the input) of the saw had not been a multiple of 10 (like 2098
    instead of 2100), because Queue1 is full and is therefore blocking the entry of additional finished short
    Enterprise Dynamics Tutorial 21
    timbers. In order to understand this, please realize that Queue1 will send short timbers to the two milling
    machines one by one).
    Based on this report, measure that 49.6 windows on average per hour have been produced!
    The option Graphs gives you a graphical representation of a variable. To be able to make a graph, you need to
    switch on the History option for the atom that you want to represent in the graph. This is done by using the
    History option in the Simulation menu.
    Warning! The option ‘General history’ (by checking) as well as the individual atom has to be selected, see
    Picture
    6‑7. At first it might seem a good idea to turn history on for all atoms, click ‘Select All atoms’. However,
    recording history this way, especially for large models, will have the effect of slowing down the simulation
    unnecessarily.
    At this stage, we will only focus on the possibility of creating graphs by means of the menu option Graphs,
    because the atom Graph often needs to be programmed. The Graph atom might be convenient for advanced
    users, because all possible variables can be defined by using 4DScript.
    Picture 6-7: History
    If the general History is switched on and some of the atoms are selected in the history window, then the run
    data of each of these atoms is collected. This data can be visualized with the Graph option in the results
    menu.
    By clicking on the option Graphs in the Results menu, a window appears. In this window you can select the
    atom of which you want to make a graphical representation. If an atom is selected and history for that specific
    Enterprise Dynamics Tutorial 22
    atom was recorded, a window such as in Picture 6‑8 appears. If no atom is selected a selector window appears.
    In that case first choose one of the atoms for which you have recorded history.
    Picture 6-8: Input window Graph
    We will first go over the last two input fields:
    Start time
    The starting time of the graph is to be entered here. If 0 is indicated, the graph will start from the
    beginning of the simulation run. If the value 10 is entered, then the graph will not include the first
    10 seconds of the run.
    Duration
    In this input field, you indicate how long the graph will be recorded. For example, if the value 3600 is
    entered here, the first hour after the starting time of the graph will be shown. The default value is
    “time”, the number of simulated seconds that have passed when the simulation run was stopped.
    The first input field, Graph Type, is where you can select the type of graph desired. There are 5 possibilities:


20. Queue Graph
    With this function, you can make a graph of the number of products present in an atom. In a Server
    atom, the number of products will as a rule never be higher than 1, but in a Queue atom, the queue is
    being conveyed in the graph. Picture 6‑9 represents a graph of the first buffer from the carpenter’s
    factory case study, measured between 2 and 4 simulated hours after the start of the simulation.
    Picture 6‑9: Queue graph


21. Queue Histogram
    The percentage of time each ‘storage level’ takes up is reflected here.
    Enterprise Dynamics Tutorial 23
    See Picture 6‑10, displaying the distribution of different queue lengths of Queue1 measured between 2
    and 4 hours after the start of the simulation.
    Picture 6‑10: Queue histogram


22. Status Pie
    This function enables you to represent the status of an atom in a pie chart. Picture 6‑11 shows a Status
    Pie representing the status of the first milling machine from the carpenter’s factory case study,
    measured during the first two simulated hours.
    Picture 6‑11: Status Pie


23. Status Bar
    With the option Status Bar, the status of an atom is reflected in a bar chart. See Picture 6‑12, which
    corresponds to the measurements displayed in Picture 6‑11.
    Enterprise Dynamics Tutorial 24
    Picture 6‑12: Status Bar


24. Wait Histogram
    The Wait Histogram function produces a histogram about a products waiting time in an atom. See
    Picture 6‑13, which displays the distribution of the waiting times of the short timbers in Queue1 during
    the first two simulated hours.
    Picture 6‑13: Wait Histogram


25. Setting up an experiment with the Experiment Wizard
    The three above mentioned methods for measuring results are particularly useful to display results directly
    during the simulation. For drawing conclusions about the results of the model, these techniques are less
    appropriate. In this case, an experiment is the appropriate method to use. An experiment can be created with
    the Experiment Wizard, which can be opened from the Simulate page of the main menu.
    It is important to understand the major differences between examining the results of a single simulation run
    with for example the Summary report, the results atoms or the Graphs option on one hand and
    Experimentation on the other hand: We use the first options for building and testing our model and for getting
    first results on single runs, while Experimentation is used later on in the process when we more or less trust
    Enterprise Dynamics Tutorial 25
    our model and want results on multiple runs, because we can’t draw conclusions from the outcome of a single
    run.
    Case study assignment (continued)
    Questions and assignments


26. Design an experiment with a warm-up period of 10 hours and a measurement period of 100 hours, in
    which the average window production per unit of time (hour, day or week or simulation period) can be
    determined with a 95% reliability.
    Does it correspond to your calculations made beforehand and to the results gained with the other
    techniques to measure results?
    We will demonstrate how to solve this assignment step by step.
    Basically there are four steps in experimentation:
    Step 1 Definition of the Experiment Settings
    Defines experiment settings such as the number of runs, the length of the observation period, the
    warm-up period, etcetera.
    Step 2 Definition of the Performance Measures
    Define the performance measures of atoms or a group of atoms.
    Step 3 Experimentation
    Running the experiment according to its definition. This step requires no action from the user!
    Step 4 Report definition and analysis
    If the sample rate of your monitor atom is still set to 60 seconds, or even worse 5 seconds, this might be a
    good time to set it to 36000 seconds (10 hours), or even 1E9 (1 billion seconds, about 32 years), since you will
    not need the monitor atom during the experiment. It will make your model run considerably faster during the
    experiment. Click on Experiment Wizard on the simulation page of the main menu, a window appears, see
    Picture 6-14. Start with giving the required values for the observation period, number of runs and the warm-up
    period in the Experiment Setting, see Picture 6‑14. If you are finished you can click Next.
    We are now ready to define the Performance Metrics (PFMs). In the ‘Edit Category Window’ click Add to create
    a new performance metric, see Picture 6-15. We are interested in the capacity per hour of the production
    system. We can measure this by measuring the number of products that enter the Sink. The output of the
    system can thus be measured by measuring the input of the Sink. To define a PFM on the Sink first click ‘Select
    Atom’ and select the Sink. We still need to define what we would like to measure at the Sink. Therefore click
    Add in the Performance Metric section. We can now enter a name and select a performance metric from the
    list.
    Enterprise Dynamics Tutorial 26
    Picture 6-14: Experiment Settings
    Define a PFM on the Sink named `Produced per hr’, see Picture 6‑15. The input during the 100 hours is divided
    by
    100 to get the number of windows per hour. Make sure you select the “[USER DEFINED]” option for your PFM,
    and do not forget to remove the string “[USER DEFINED] “, and type Input(cs)/100. cs in the 4DScript code
    for the current selection, (which can also be a group of atoms).
    If both the name and the PFM are entered click Ok. Click Add again to define another PFM on the Sink. In this
    case we only need a single PFM thus we can click again on Ok . We can click Add again if we need to define
    another PFM on a different atom. Click Next after all PFM have been entered.
    Enterprise Dynamics Tutorial 27
    Picture 6-15: Definition of the PFM
    The actual running of the model –after start experiment- is done in step 3, see Picture 6‑16 and may take some
    time: In every run of 100 hours about 500 windows have to be made from 2000 short timbers and 200 long
    timbers!
    Picture 6-16: Experiment Settings
    After running the experiment it’s easy to make a report, see Picture 6‑17. From this table the results can be
    exported to Excel (click ‘save as .csv’ ) or put in a report via QuickReport (click ‘Preview’).
    Enterprise Dynamics Tutorial 28
    Picture 6-17: The Results Table
    This concludes our example about experimenting. For a thorough examination of what is possible we refer to
    our document on experimentation, see Help | Tutorials.
    The model with this experiment is supplied under the name timber2.mod
    Enterprise Dynamics Tutorial 29
    7 Playing with Strategies
    In this chapter, the emphasis lies on the pre-defined strategies with which products get access to a next atom
    (input strategy), are selected out of a queue (queue discipline) and are re-directed after processing (send to).
    To demonstrate this, a simple model has been built. It consists of three Sources, three Locks (an atom
    regulating the products’ supply), five Servers, four Queues and one Switch, which is a self-built atom
    (Enterprise Dynamics offers this flexibility). The purpose of this atom is to switch the Servers on and off. There
    is no Sink; consequently all products remain in the last Queue, see also Picture 7‑1.
    The model is supplied under the name strategy.mod.
    Picture 7-1: The layout of strategy.mod
    Warning: Because the source code of the last queue atom has been adjusted and a new designed atom is used,
    beginners will not be able to build this model on their own. Advanced users can however build this model by
    using the standard atom and adjusting the code on the 2D Draw event handler on the last queue so that it can
    show up to 100 products in it. The atom behavior can be changed from the Atom Editor. Because the Switch is
    also not a standard atom, advanced users will have to switch the servers on and off manually. This is possible
    with the Availability Control atoms, among others.
    From the three sources, three types of customers arrive at the queues. Their arrival is regulated in such a way
    that the three queues will get filled rather quickly. We have set the option ‘Number of products’ on each of
    the sources to allow only 10 of each type of customer to leave the source. From the queue, the customers are
    helped by a server and after that end up in the last queue. The role of the switch is to switch the servers on
    and off.
    In principle, all servers can be reached from the three queues, the first server can always be reached through
    channel 1, the second through channel 2, etc. However, in the original settings of strategy.mod, only server 1
    has been switched on (by means of the switch). As a result, during a simulation, the first server is colored
    green while the others turn red. In addition, the default parameters are applied, which means that:


27. all input strategies are still on Any input channel,


28. all queue disciplines are still on First in first out,


29. all send to on the Queues are set to 4. A random open channel: choose a random channel from all the
    open output channels. This means every Server available can be the next station from a Queue.
    The products and their icons are visible in the last queue. As a result, the order in which the products have
    entered the queue is accurately visible (provided the queue discipline is set on FIFO).
    Because the products receive a label when leaving the source, they can also be sorted out in the last queue on
    the basis of this label. Labels can be thought of as tags hanging on a product. They can represent a color, a
    weight, a size or something similar. They are defined in 4DScript, Enterprise Dynamics’ programming language,
    and play an important role in most models.
    Enterprise Dynamics Tutorial 30
    Look at the other parameters of strategy.mod and carry out a few simulation runs to find out more about it.
    After that, we are now really ready for experimenting with strategies…
    7.1 Adjusting the input strategy
    The input strategy regulates the access to an atom from preceding atoms (i.e. those with an output channel
    connected to one of its input channels). The role of the input strategy is to define the order in which products
    will be accepted from the available channels.
    You can think of the input strategy to the sequencing of traffic lights where some roads have a greater
    proportion of time at Green than more minor roads.
    The first three input strategies open all input channels and the last two open one input channel each time.
    As an exercise, we are now going to look at the effect of the input strategy change on server 1 in the unaltered
    initial model. Change the input strategy as specified below and consider what is happening to the performance
    of the systems after observing a few runs.


30. Any inputchannel
    When activated, this strategy opens all input channels of an atom. If more than one of the atoms that
    are connected via the input channel can be forwarded, the atom with the lowest number as input
    channel will have priority. As long as the products keep entering through the first channel, the other
    channels will be blocked.


31. Largest queue
    When activated, this strategy opens all input channels of an atom. If more than one of the atoms that
    are connected via the input channel can be sent, the atom with the longest queue or largest contents
    will have priority. Note that in the case of several equally long queues, the input channel with the
    lowest number is always chosen.


32. Longest waiting
    When activated, this strategy opens all input channels of an atom. If more than one of the atoms that
    are connected via the input channel can be sent, the atom with the highest waiting time will have
    priority. In the case of several atoms with an equal waiting time, the input channel with the lowest
    number is always chosen. Note that it does not mean that the queues get approximately equally long, as
    is the case in the previous option.


33. Round robin
    This strategy first opens the first input channel and then waits for a product to be sent through this
    input channel. In the second cycle, it is the turn of the second input channel etc. When the products
    have run through the last input channel, the procedure is resumed with the first one.
    This strategy becomes active after the first product! So, in case of three input channels this
    strategy gives x,2,3, 1,2,3, 1,2,3 where x can be 1,2 or 3!


34. Round robin with content check
    This strategy is similar to the Round Robin, with that distinction that when the atom connected to the
    desired input channel has no content, the next will be checked. When none of the incoming channels
    has something to hand over, the strategy will wait for the first channel that will offer something and
    will continue from there on.
    This strategy becomes active after the first product! So, in case of three input channels this
    strategy gives x,2,3, 1,2,3, 1,2,3 where x can be 1,2 or 3!


35. Channel 1.
    In this case, you can enter a specific input channel through which all products must enter. If 1 is
    entered, the products may enter only through input channel 1. Note that this rule is not valid for the
    first product entering as all channels are open initially.
    7.2 Changing the Queue Discipline
    The purpose of this section is to change the Queue Discipline of the last queue and to observe the