Page 10 - InnoRenew CoE International Conference 2020, Integrating Sustainability and Health in Buildings
P. 10
ng discrete optimization methods in decision support for structural design
Balázs Dávid
InnoRenew CoE, balazs.david@innorenew.eu / University of Primorska, balazs.david@famnit.upr.si
Structural design is a complex process of several stages that is used for the design and
development structural plans. The stages of this of this process (planning, design and detailing)
have to be performed sequentially, each stage using the output of the previous one as its input.
As these stages are complex, even separately, efficient solution methods can be useful to aid
the decision-making processes of civil engineers. Early decisions in the design process (such
as topology and material combination choices) affect the future steps and overall performance
(such as energy demand or costs). These effects are not known in advance and can only be
estimated. Providing multiple possible design suggestions by quick heuristic algorithms can
help with quantifying the effects of these early decisions.
In this presentation, we will introduce heuristic optimization methods for the design stage of
the above process. These methods use preliminary designs as their input and aim to improve
their quality through several local transformation steps. Different constraints are considered
during this process, and the solution is optimized by taking multiple different cost objectives
into account. As each local transformation step shifts from a feasible solution into another
one, multiple possible solutions are visited in the solution space. The best ones are saved and
presented by the system as possible suggestions.
To measure the quality of these resulting suggestions, we also develop a mathematical model
that is able to calculate the costwise optimal solution. The quality of our suggestions will also
be compared to this solution.
Keywords: structural design, local search, mathematical model, decision support
Acknowledgements: The author gratefully acknowledges the European Commission for funding
the InnoRenew project (Grant Agreement #739574) under the Horizon2020 Widespread-Teaming
program and the Republic of Slovenia (investment funding from the Republic of Slovenia and
the European Union’s European Regional Development Fund) and is grateful for the support of
the National Research, Development and Innovation Office - NKFIH Fund No. SNN-117879.
INNORENEW COE INTERNATIONAL CONFERENCE 2020
10
Balázs Dávid
InnoRenew CoE, balazs.david@innorenew.eu / University of Primorska, balazs.david@famnit.upr.si
Structural design is a complex process of several stages that is used for the design and
development structural plans. The stages of this of this process (planning, design and detailing)
have to be performed sequentially, each stage using the output of the previous one as its input.
As these stages are complex, even separately, efficient solution methods can be useful to aid
the decision-making processes of civil engineers. Early decisions in the design process (such
as topology and material combination choices) affect the future steps and overall performance
(such as energy demand or costs). These effects are not known in advance and can only be
estimated. Providing multiple possible design suggestions by quick heuristic algorithms can
help with quantifying the effects of these early decisions.
In this presentation, we will introduce heuristic optimization methods for the design stage of
the above process. These methods use preliminary designs as their input and aim to improve
their quality through several local transformation steps. Different constraints are considered
during this process, and the solution is optimized by taking multiple different cost objectives
into account. As each local transformation step shifts from a feasible solution into another
one, multiple possible solutions are visited in the solution space. The best ones are saved and
presented by the system as possible suggestions.
To measure the quality of these resulting suggestions, we also develop a mathematical model
that is able to calculate the costwise optimal solution. The quality of our suggestions will also
be compared to this solution.
Keywords: structural design, local search, mathematical model, decision support
Acknowledgements: The author gratefully acknowledges the European Commission for funding
the InnoRenew project (Grant Agreement #739574) under the Horizon2020 Widespread-Teaming
program and the Republic of Slovenia (investment funding from the Republic of Slovenia and
the European Union’s European Regional Development Fund) and is grateful for the support of
the National Research, Development and Innovation Office - NKFIH Fund No. SNN-117879.
INNORENEW COE INTERNATIONAL CONFERENCE 2020
10
