The model of a system, following the process-oriented approach
is composed of two kinds of information: the flowchart diagram and
the experimental data. Different icons identify the flowchart and
data modules in the BasicProcess package.
In this document the construction of a new model, with both
components, will be detailed.
The first step to construct a new model is to create a new Modelica model that extends the ARENALib.Draft model.
model tutorial extends DESLib.ARENALib.Draft; equations end tutorial;
The flowchart diagram represents the possible flows of entities in the system. These flows can be described in ARENALib using the flowchart modules included in the BasicProcess package.
The flow of entities must start in one or more Create modules, and usually end in a Dispose modules.
In order to construct the flowchart diagram, the modules have to
be included from the BasicProcess package into the diagram of the
newly created model.
After that, the ports of the included blocks have to be properly
connected.
The flowchart diagram of a single server with queue system is shown
in the next figure.
The parameters of each module has to be defined to represent the behavior of the system (e.g. the inter-arrival times, the processing delays, etc.).
If an algebraic loop
is inserted in the diagram (i.e. the output of one module is connected to a previous module in the flowchart diagram, creating a loop), it can be broken using the BreakLoop model of the DEVSLib library.To complete the development of the model, the experimental
data has to be included.
This can be performed in ARENALib by inserting data modules from
the BasicProcess package into the diagram of the model.
The data modules required by the single server with queue model are shown in the next figure.
The parameters of each data module has also to be defined to represent the behavior of the system (e.g. capacity of the resource, etc.)
In ARENALib, each flowchart module generates results for the
simulation run by calculating predefined statistical indicators.
These indicators include counters, discrete and time-dependent
statistical information. The documentation of each module describes
the statistical indicators calculated by each module.
By default, all the modules write their results to the
ARENALIB_RESULTS.txt file.
However, more statistical indicators can be calculated including Record modules in the models. The Record module allows to calculate counters and tally statistical indicators.
Also, the evolution of some of the variables of the model
over the time can be plotted using the standard functionalities
provided by Dymola.
For example, the number of entities in a queue can be observed with
the variable Process.Queue.NQ, or the number of busy resource units
with Resource.NR.