10
2 Simulation-Based Optimization
chapter 5 to explain the relationship of the individual model elements of different
simulation methods in detail. To identify the optimal production setup, optimization will be used to increase the energy efficiency in production (section 2.2).
“Advances in the field of metaheuristics -the domain of optimization that augments traditional mathematics with artificial intelligence and methods based on
analogs to physical, biological or evolutionary processes- have led to the creation
of optimization engines that successfully guide a series of complex evaluations
with the goal of finding optimal values for the decision variables” [Ap+2004,
p. 76]. Thus, optimization engines appear to be the ideal tool for dealing with the
issue of increasing resource efficiency in production plants. Therefore, chapter 2
briefly describes relevant terms and definitions in simulation and optimization as
well as possible combinations of the two. Reference to advanced literature will
be made in the appropriate places of the text.
2.1
Modeling and Simulation
In a broad sense, simulation refers to the process of constructing, using and verifying a virtual, mostly computer-based experiment or reproduction of a system to
gain insights on the system that are transferable to reality [Ba+2005, pp. 3–4;
Du2018, p. 13]. A simulation study aims at the abstract and consistent imitation
of a real-world system with its dynamic processes in an experimental computer
model that contains a variation of parameters and structures, constituting a system’s behavior over time [Ba1998, p. 3; VD2014a, p. 3] to conduct “experiments
with this model for the purpose either of understanding the behavior of the system
or of evaluating various strategies (within the limits imposed by a criterion or set
of criteria) for the operation of the system” [Sh1975, p. 2]. A system consists of
a group of objects, which are characterized by attributes, that interact with each
other and have interdependencies towards the accomplishment of a given purpose
within defined system boundaries [Ba+2005, p. 8]. A simulation allows to run
through numbers of scenarios and various system settings that cannot easily be
altered in a real production as this would be too costly and disruptive. The process
of incrementally and interactionally varying model parameters and comparing the
system’s behavior against a starting situation is referred to as simulation in the narrow sense [Du2018, p. 13]. Assumptions about the systems can be considered, as
they are sometimes necessary to develop a model [Go2015, p. 15]. Consequently,
different ways to study a system can be followed (Figure 2.1).
To perform a simulation, it is required to have a look at a typical simulation life
cycle and its stages. Every simulation study starts with the problem formulation
2 Simulation-Based Optimization
chapter 5 to explain the relationship of the individual model elements of different
simulation methods in detail. To identify the optimal production setup, optimization will be used to increase the energy efficiency in production (section 2.2).
“Advances in the field of metaheuristics -the domain of optimization that augments traditional mathematics with artificial intelligence and methods based on
analogs to physical, biological or evolutionary processes- have led to the creation
of optimization engines that successfully guide a series of complex evaluations
with the goal of finding optimal values for the decision variables” [Ap+2004,
p. 76]. Thus, optimization engines appear to be the ideal tool for dealing with the
issue of increasing resource efficiency in production plants. Therefore, chapter 2
briefly describes relevant terms and definitions in simulation and optimization as
well as possible combinations of the two. Reference to advanced literature will
be made in the appropriate places of the text.
2.1
Modeling and Simulation
In a broad sense, simulation refers to the process of constructing, using and verifying a virtual, mostly computer-based experiment or reproduction of a system to
gain insights on the system that are transferable to reality [Ba+2005, pp. 3–4;
Du2018, p. 13]. A simulation study aims at the abstract and consistent imitation
of a real-world system with its dynamic processes in an experimental computer
model that contains a variation of parameters and structures, constituting a system’s behavior over time [Ba1998, p. 3; VD2014a, p. 3] to conduct “experiments
with this model for the purpose either of understanding the behavior of the system
or of evaluating various strategies (within the limits imposed by a criterion or set
of criteria) for the operation of the system” [Sh1975, p. 2]. A system consists of
a group of objects, which are characterized by attributes, that interact with each
other and have interdependencies towards the accomplishment of a given purpose
within defined system boundaries [Ba+2005, p. 8]. A simulation allows to run
through numbers of scenarios and various system settings that cannot easily be
altered in a real production as this would be too costly and disruptive. The process
of incrementally and interactionally varying model parameters and comparing the
system’s behavior against a starting situation is referred to as simulation in the narrow sense [Du2018, p. 13]. Assumptions about the systems can be considered, as
they are sometimes necessary to develop a model [Go2015, p. 15]. Consequently,
different ways to study a system can be followed (Figure 2.1).
To perform a simulation, it is required to have a look at a typical simulation life
cycle and its stages. Every simulation study starts with the problem formulation