Abstract

With the growth in the criticality of buildings’ lifecycle performance, building performance simulation (BPS) is becoming a more prominent step in the building design process. BPS’s ability to approximate the performance of a building in the real world enables BPS to be used for ensuring compliance and trade-off of design parameters at a late period of design. The quantitative information provided by BPS, if applied at an earlier period of design, has the potential to assist in more impactful decisions (which are currently being based on rules of thumb). The problem, however, is that the lack of details set at such early design translates to a very large number of scenarios to be simulated, requiring extensive time and computational power that is not available to designers during that phase.
To try limiting the number of scenarios to be simulated, the main goal of this study is to provide a deeper understanding of the impact caused by building design parameters and building characteristics. To accomplish that, data analyses were performed on a database of representative building models to investigate the sensitivity of outputs (Energy use intensity – EUI and net present worth of cost – NPW) to design parameters (architectural, electrical and mechanical systems) and the sensitivity of parameters’ impact to building models. For each building model, energy simulations were performed based on a one-parameter-at-a-time (OAT) sampling, and the costs were evaluated through developed cost models.
The results of this study show that wall and roof insulation, window type, window-to-wall ratio, and lighting efficiency parameters are sensitive to the analyzed model. When analyzing different building groups (e.g. low- and high-rise) separately, it was found that parameters’ significance is correlated to building characteristics (e.g. building height). This can be particularly of extreme importance for limiting design alternatives at the early stage of design when the multiplicity of design scenarios is currently limiting the applicability of BPS in the early-stage decision-making for building designs. In the future, the use of more advanced data analysis tools will help improve the accuracy of the observed results as well as provide an inclusive classification for the level of impact of various design parameters in different building types.