196
7 Summary and Outlook
energy aspects in production simulations. Chapter 4 therefore concluded with
a derivation of the current research demand. The research gaps identified were
the lack of an easy applicability of approaches, the missing depiction of dynamic interdependencies of the discrete material flow and the continuous energy
consumption, lacking support of integrated optimization, and missing approaches
solving the combined depiction of production details and energy aspects in one
single software solution. The identified research gaps were concretized into requirements and objectives in chapter 5, followed by the development of a conceptual
framework for the simulation-based optimization of energy efficiency in production which was implemented in a fictional case scenario at the end of the chapter.
In chapter 6, the developed methodology has been tested on a practical example.
In addition to details on data acquisition, processing and validation, the simulation
model and the optimization experiments using differently resolved energy data
have been described. Additionally, the necessity of combined simulation approaches in practice has been examined. Chapter 6 closed with a financial evaluation of
the optimization potential in the practical use case and an analysis of the practical
applicability of the simulation-based optimization methodology.
The methodology consists of two modules, a simulation and an optimization
module. The simulation module is divided into three components in order to
represent the production processes at all levels, from the internal process level
to the macro level. Following the industry standard in production simulation,
a material flow component is defined in which the processes in production are
modeled discretely in a process-oriented manner. In order to exactly reproduce the
machine processes with their individual machine states in production, an agentbased machine logic was developed to represent the processes at the micro level.
Thus, the machine states, state changes, and state durations can be modeled realistically. The link to the energy consumption behavior is established via the energy
component of the simulation module. The energy component comprises energy
load profiles which are assigned to the individual machine states and are used
accordingly during the simulation experiment runs.
In order to not only model the energy aspects in production but also to use
them for optimization scenarios, lexicographic objective functions have been derived that propose ideal parameter configurations for the energy-efficient operation
of the production line using simulation-based optimization experiments. The focus
of the optimization is on the reduction of the total energy consumption by avoiding non-value adding machine states. In addition, the methodology includes the
possibility to determine an offset parameter to change the machine starts within an
allowed time frame to reduce occurring load peaks. The developed methodology
was first implemented in a fictitious case study using the software AnyLogic (PLE
7 Summary and Outlook
energy aspects in production simulations. Chapter 4 therefore concluded with
a derivation of the current research demand. The research gaps identified were
the lack of an easy applicability of approaches, the missing depiction of dynamic interdependencies of the discrete material flow and the continuous energy
consumption, lacking support of integrated optimization, and missing approaches
solving the combined depiction of production details and energy aspects in one
single software solution. The identified research gaps were concretized into requirements and objectives in chapter 5, followed by the development of a conceptual
framework for the simulation-based optimization of energy efficiency in production which was implemented in a fictional case scenario at the end of the chapter.
In chapter 6, the developed methodology has been tested on a practical example.
In addition to details on data acquisition, processing and validation, the simulation
model and the optimization experiments using differently resolved energy data
have been described. Additionally, the necessity of combined simulation approaches in practice has been examined. Chapter 6 closed with a financial evaluation of
the optimization potential in the practical use case and an analysis of the practical
applicability of the simulation-based optimization methodology.
The methodology consists of two modules, a simulation and an optimization
module. The simulation module is divided into three components in order to
represent the production processes at all levels, from the internal process level
to the macro level. Following the industry standard in production simulation,
a material flow component is defined in which the processes in production are
modeled discretely in a process-oriented manner. In order to exactly reproduce the
machine processes with their individual machine states in production, an agentbased machine logic was developed to represent the processes at the micro level.
Thus, the machine states, state changes, and state durations can be modeled realistically. The link to the energy consumption behavior is established via the energy
component of the simulation module. The energy component comprises energy
load profiles which are assigned to the individual machine states and are used
accordingly during the simulation experiment runs.
In order to not only model the energy aspects in production but also to use
them for optimization scenarios, lexicographic objective functions have been derived that propose ideal parameter configurations for the energy-efficient operation
of the production line using simulation-based optimization experiments. The focus
of the optimization is on the reduction of the total energy consumption by avoiding non-value adding machine states. In addition, the methodology includes the
possibility to determine an offset parameter to change the machine starts within an
allowed time frame to reduce occurring load peaks. The developed methodology
was first implemented in a fictitious case study using the software AnyLogic (PLE