7.3 Outlook and Future Work
205
extension of the methodology to depict energy consumption of the technical building equipment could be considered as an option. Cooling and heating systems,
air conditioning and other energy consuming equipment of the building infrastructure could be modeled to be able to assess and to optimize the share of energy
consumption they cause. The simulation-based optimization methodology grants
flexibility which overcomes the need for predefined system borders. Besides the
production processes, logistics processes, as well as technical building equipment
can be modeled with the simulation paradigms used. A limitation to the production area is therefore not necessary. It should be examined whether the creation
of further simulation modules for uniform modeling of the building infrastructure
would simplify the creation of the simulation model, as the defined machine logic
types of the machine component might have to be changed to represent heating
and cooling processes or transportation systems realistically.
The cooperation with the practice partner repeatedly led to the demand for
the use of the methodology to model compressed air consumptions as well as
cooling and lubricant circuits. Initial research and experiments have shown that
the hybrid approach could also be suitable for depicting such media, as they also
exhibit continuous behavior which is highly dependent on machine behavior.
So far, all required data has been collected and matched manually. The automated data integration would bear potential for improvement. An interface to the
required data sources such as the ERP system for production and process data as
well as a coupling to the energy data portal could provide the latest data and thus
improve the data quality used for the simulation model. In order to provide a better and faster decision support, the automated transfer of optimization parameter
values without manual interference to the production machines should be tested.
In addition to the different approaches for extending and improving the simulation model, the optimization component of the methodology also offers starting
points for future work. Whereas in this approach only a parameter optimization
was carried out in two separate optimization experiments, a next step could be
an optimization based on a cost function. The definition of a cost function would
allow the overall optimization purely under consideration of the energy costs focusing the total consumption and the peak optimization costs in parallel. Due to the
lack of a reliable database for the grid costs, the creation of a cost function has so
far been dispensed. The introduction of a cost function would additionally allow
the consideration of time-dependent energy prices in the course of the day as
proposed in Johannes, Wichmann, and Spengler [JWS2019].
One option to overcome the limitation of the methodology to simulate and
optimize only known and measured machine consumptions might be the deep
205
extension of the methodology to depict energy consumption of the technical building equipment could be considered as an option. Cooling and heating systems,
air conditioning and other energy consuming equipment of the building infrastructure could be modeled to be able to assess and to optimize the share of energy
consumption they cause. The simulation-based optimization methodology grants
flexibility which overcomes the need for predefined system borders. Besides the
production processes, logistics processes, as well as technical building equipment
can be modeled with the simulation paradigms used. A limitation to the production area is therefore not necessary. It should be examined whether the creation
of further simulation modules for uniform modeling of the building infrastructure
would simplify the creation of the simulation model, as the defined machine logic
types of the machine component might have to be changed to represent heating
and cooling processes or transportation systems realistically.
The cooperation with the practice partner repeatedly led to the demand for
the use of the methodology to model compressed air consumptions as well as
cooling and lubricant circuits. Initial research and experiments have shown that
the hybrid approach could also be suitable for depicting such media, as they also
exhibit continuous behavior which is highly dependent on machine behavior.
So far, all required data has been collected and matched manually. The automated data integration would bear potential for improvement. An interface to the
required data sources such as the ERP system for production and process data as
well as a coupling to the energy data portal could provide the latest data and thus
improve the data quality used for the simulation model. In order to provide a better and faster decision support, the automated transfer of optimization parameter
values without manual interference to the production machines should be tested.
In addition to the different approaches for extending and improving the simulation model, the optimization component of the methodology also offers starting
points for future work. Whereas in this approach only a parameter optimization
was carried out in two separate optimization experiments, a next step could be
an optimization based on a cost function. The definition of a cost function would
allow the overall optimization purely under consideration of the energy costs focusing the total consumption and the peak optimization costs in parallel. Due to the
lack of a reliable database for the grid costs, the creation of a cost function has so
far been dispensed. The introduction of a cost function would additionally allow
the consideration of time-dependent energy prices in the course of the day as
proposed in Johannes, Wichmann, and Spengler [JWS2019].
One option to overcome the limitation of the methodology to simulate and
optimize only known and measured machine consumptions might be the deep