R&D Engineer, EnginSoft SpA
Silvia Demattè has an MSc in Building Engineering from Politecnico of Milan with a thesis on energy performance (measurements and calculations). She works in the R&D Department of EnginSoft, an Italian-based multinational consulting company active in the field of Simulation-Based Engineering and Sciences. She has been involved in the BENIMPACT research project, where she developed methodologies and software tools for the optimization of sustainable buildings, and on the DIRECTION project, whose aim is to demonstrate innovative, effective solutions for very low energy new buildings.
Optimization implementation in the construction industry is still quite rare, and design practitioners usually follow general rules-of thumbs. The main reason can be identified as a lack in tools, experience and knowledge required for its implementation.
A real structural optimization test-case provided by Studio Iorio (Bergamo, Italy) is used to highlight and analyze benefits and barriers. It is demonstrated that the adoption of automated optimization in the building industry is a worthwhile endeavor even if there are important issues to be addressed.
First of all, the classical design process is heavily influenced by architectural constraints of space and form especially in earlier design stages, and decisions taken are often based on previous experience or intuition but do not have a rigorous justification. Indeed, optimization is approached as a manual iterative trial and error approach to find structurally or financially feasible solutions, and it is seen as a very complicated (prohibitive) practice requiring a long time for modelling and method customization.
Moreover the interdisciplinary nature of building design, with input from clients, architects and engineers, complicates the process and leads to numerous iterations and revisions. Thus a project may change rapidly, and the optimization effort may immediately become obsolete.
In such a scenario, an integrated, parametric and a multi-objective design environment could really simplify, rationalize, economize and provide rigor to the entire process. Indeed it enables generative form-making and form-finding and provides rules to compare design alternatives on the basis of performance metrics.
A driver to push integration of parametric modeling and optimization into the construction industry can be identified in BIM (Building Information Modelling), which provides methods and tools to represent the building as an integrated database of coordinated information from which it is possible to automatically generate input files for various performance simulation tools. This allows practitioners to efficiently generate and modify building models and simulate and analyze multiple performances (i.e. energetic, structural, acoustic, lightning, etc.), leading to a more comprehensive exploration of the design space and providing better decision support for the stakeholders.
A constant use of appropriate design, analysis and optimization techniques for routine tasks develops skills and increases efficiency by saving design time, avoiding the need to iterate a design by hand to find structurally or financially feasible solutions, and reducing the tedium of routine tasks. Rapid generation and evaluation of a large number of design alternatives, with better exploration of the design space (compared to what can be achieved manually) can be strongly appreciated in early design stages when it is also possible to discover unexpected feasible solutions within the defined constraints. Moreover, if the context changes in a later design stage, parametric modeling allows objects to be automatically updated.
Financial savings are an obvious potential driver for optimization methods. For example, cost can be simplistically equated to structural weight or piece-count and detailed connections can be also considered. Furthermore, maximization of floor space and indoor quality will control the potential revenue for offices and residential buildings, influencing the financial feasibility of a project as a whole.
However practical economic concerns include the cost of software licenses, especially when applied to a single project, and many commercial tools seem to be targeted to aerospace and automotive industries and to be inaccessible to designers who do not use them on a regular basis as they require technical and theoretical expertise.
The fact that initial investments are needed to develop capabilities before cost-benefits are achieved suggests that optimization can first be offered to the building industry as a specialized service to become a more common practice in the following years.