7.2 Critical Appraisal of the Methodology
197
8.5.1), before it was tested in industrial practice using the example of two production lines at an automotive supplier in chapter 6. The implementation shows that
the developed methodology is working for practical applications, but the quality
of the results is highly reliant on the quality of the database and the completeness
of information from the real production used for the construction of the simulation
model.
7.2
Critical Appraisal of the Methodology
The objective of the presented approach is the depiction of combined material
and energy flow simulation in production to create a realistic model showing
the dynamic behavior of the energy consumption in production process. The use
of a multi-method simulation software simplifies the hybrid modeling process.
The complexity of interface management, complex data exchanges, and synchronization requirements are reduced compared to the usage of different software
packages for the single simulation paradigms. Occurring peak demands and timecontinuous energy consumption can be shown closer to reality compared to DES
models based on measured operating states, which are considered to be constant
over a defined period of time.
The practical implementation has shown that it is possible to build a hybrid
simulation model, using the three modeling paradigms SD, DES, and ABS for
representing the energy consumption behavior in production on the basis of historical data. It was also pointed out that the simulation model, in combination
with forecast figures regarding quantities and planned schedules, can be used to
depict the future consumption very precisely with its upcoming peak loads and
non-value-adding production phases. From the optimization experiments carried
out, alternative courses of action for the energy-efficient control of machine states
emerge which lead to energy consumption reductions of 10% and more in production situations similar to the fictional example and savings of still 6% in strongly
connected production lines. The level of optimization potential is strongly dependent on the starting point. If measures to increase efficiency have already been
implemented in production, e.g., rough specifications have been installed that
machines must be switched off during downtimes exceeding 60 or 90 minutes,
the optimization potential is correspondingly lower. If machines are not decoupled from one another by buffers, it makes sense to take a holistic view of the
production line. In this case again, the optimization potential will be lower than
for machine state optimization, in which the machines can be individually set to
their optimum.
197
8.5.1), before it was tested in industrial practice using the example of two production lines at an automotive supplier in chapter 6. The implementation shows that
the developed methodology is working for practical applications, but the quality
of the results is highly reliant on the quality of the database and the completeness
of information from the real production used for the construction of the simulation
model.
7.2
Critical Appraisal of the Methodology
The objective of the presented approach is the depiction of combined material
and energy flow simulation in production to create a realistic model showing
the dynamic behavior of the energy consumption in production process. The use
of a multi-method simulation software simplifies the hybrid modeling process.
The complexity of interface management, complex data exchanges, and synchronization requirements are reduced compared to the usage of different software
packages for the single simulation paradigms. Occurring peak demands and timecontinuous energy consumption can be shown closer to reality compared to DES
models based on measured operating states, which are considered to be constant
over a defined period of time.
The practical implementation has shown that it is possible to build a hybrid
simulation model, using the three modeling paradigms SD, DES, and ABS for
representing the energy consumption behavior in production on the basis of historical data. It was also pointed out that the simulation model, in combination
with forecast figures regarding quantities and planned schedules, can be used to
depict the future consumption very precisely with its upcoming peak loads and
non-value-adding production phases. From the optimization experiments carried
out, alternative courses of action for the energy-efficient control of machine states
emerge which lead to energy consumption reductions of 10% and more in production situations similar to the fictional example and savings of still 6% in strongly
connected production lines. The level of optimization potential is strongly dependent on the starting point. If measures to increase efficiency have already been
implemented in production, e.g., rough specifications have been installed that
machines must be switched off during downtimes exceeding 60 or 90 minutes,
the optimization potential is correspondingly lower. If machines are not decoupled from one another by buffers, it makes sense to take a holistic view of the
production line. In this case again, the optimization potential will be lower than
for machine state optimization, in which the machines can be individually set to
their optimum.