4.1 Selection and Evaluation of Relevant Research Approaches
79
coupling of two simulation models executed in their own simulation system but
through the use of the Discrete Event System and Differential Equation System
Specification (DEV & DESS) approach from the field of systems theory. Schmidt
and Pawletta use the discrete event simulator MATLAB/SimEvents without
enabling the continuous model library to model their transaction-oriented process
chains. To calculate the resource consumption of machines, a newly developed
model library is used. It contains specific manufacturing processes and can submit
its calculation results in form of state variable vectors to the simulation software.
The hybrid simulation model consists of three parts, the material flow component,
the state-based controller and a subsystem for displaying time which are all validated using the example of a hardening furnace. The results show that the oven
reaches the preset temperature but the set time cannot be met. The approach can
be used to validate technical data, to gain information for a proactive maintenance
of machines as well as for the reduction of shut down times.
In another paper in 2017, Pawletta and Schmidt together with Junglas
present a multi-modeling approach, combining several modeling methods, such as
discrete event modeling for the material flow, state graphs for the process control
as well as continuous models for the process physics, to describe and investigate
the dynamic system behavior of a production line [PSJ2017]. The multi-modeling
approach subdivides a manufacturing system in three general layers and allows
the implementation of different models with varying levels of detail, organized
in a library. Thus, it supports the component-oriented and flexible refinement
of production line models. Pawletta, Schmidt, and Junglas illustrate their
approach using MATLAB/Simulink to model a furnace component and its timerelated energy consumption behavior in a set of models with different levels of
abstraction and varying levels of detail. They compare refinement costs, simulation runtimes and the level of accuracy and state that the energy results among
the different model accuracies vary by 10–15% [PSJ2017, p. 122].
Schlüter et al. use a bidirectional coupling of material flow and energy
models created in a hybrid simulation environment based on MATLAB, Simulink,
and Stateflow, supporting continuous and time discrete processes [Sc+2017]. The
coupled simulation model communicates with a process control model, the process control receives system and process parameters as well as plan data from the
simulation model and returns order and control data to the simulation model for
validation. On the basis of the returned configurations, various operating scenarios
and the effectiveness of individual optimization measures can be simulated and
evaluated. Schlüter et al. validate their approach comparing simulation results
to measured reference data from a non-ferrous melting and die-casting plant.
79
coupling of two simulation models executed in their own simulation system but
through the use of the Discrete Event System and Differential Equation System
Specification (DEV & DESS) approach from the field of systems theory. Schmidt
and Pawletta use the discrete event simulator MATLAB/SimEvents without
enabling the continuous model library to model their transaction-oriented process
chains. To calculate the resource consumption of machines, a newly developed
model library is used. It contains specific manufacturing processes and can submit
its calculation results in form of state variable vectors to the simulation software.
The hybrid simulation model consists of three parts, the material flow component,
the state-based controller and a subsystem for displaying time which are all validated using the example of a hardening furnace. The results show that the oven
reaches the preset temperature but the set time cannot be met. The approach can
be used to validate technical data, to gain information for a proactive maintenance
of machines as well as for the reduction of shut down times.
In another paper in 2017, Pawletta and Schmidt together with Junglas
present a multi-modeling approach, combining several modeling methods, such as
discrete event modeling for the material flow, state graphs for the process control
as well as continuous models for the process physics, to describe and investigate
the dynamic system behavior of a production line [PSJ2017]. The multi-modeling
approach subdivides a manufacturing system in three general layers and allows
the implementation of different models with varying levels of detail, organized
in a library. Thus, it supports the component-oriented and flexible refinement
of production line models. Pawletta, Schmidt, and Junglas illustrate their
approach using MATLAB/Simulink to model a furnace component and its timerelated energy consumption behavior in a set of models with different levels of
abstraction and varying levels of detail. They compare refinement costs, simulation runtimes and the level of accuracy and state that the energy results among
the different model accuracies vary by 10–15% [PSJ2017, p. 122].
Schlüter et al. use a bidirectional coupling of material flow and energy
models created in a hybrid simulation environment based on MATLAB, Simulink,
and Stateflow, supporting continuous and time discrete processes [Sc+2017]. The
coupled simulation model communicates with a process control model, the process control receives system and process parameters as well as plan data from the
simulation model and returns order and control data to the simulation model for
validation. On the basis of the returned configurations, various operating scenarios
and the effectiveness of individual optimization measures can be simulated and
evaluated. Schlüter et al. validate their approach comparing simulation results
to measured reference data from a non-ferrous melting and die-casting plant.