THE EFFECT OF MODELER DECISIONS ON SIMULATION UNCERTAINTY : SOME IMPLICATIONS FOR USER INTERFACE DESIGN

Previous research has shown that building energy simulation (BES) users introduce uncertainty to the simulation process. Analogous research in the area of computer security provides a framework for identifying BES user interface (UI) features that may be modified to reduce differences in simulation results caused by modelers. An original data set from participant energy modelers is expanded through Monte Carlo sampling and then analyzed with random forests. The Monte Carlo data is compared to the factor analysis data from a previous paper to uncover basic classes of impact for modeler decision categories. The results indicate that UI features that require complicated input methods (e.g. HVAC and lighting power entry) or that demand specific estimates of unknowns at the time of modeling (e.g. plug loads and schedule) are the most significant sources of output differences. Designs for BES UI features are provided that are expected to significantly reduce uncertainty introduced by the modeler. For cases where the suggested changes to the UI are non-trivial, a methodology for defining changes is proposed. These UI designs will both rely on and inform future Uncertainty Quantification (UQ) research in BES.

[1]  Nicholas Long,et al.  The Building Component Library: An Online Repository to Facilitate Building Energy Model Creation , 2012 .

[2]  Jlm Jan Hensen,et al.  Uncertainty analysis for conceptual building design - a review of input data , 2006 .

[3]  Huafen Hu,et al.  Risk-conscious design of off-grid solar energy houses , 2009 .

[4]  Sébastien Lê,et al.  FactoMineR: An R Package for Multivariate Analysis , 2008 .

[5]  Christiaan J. J. Paredis,et al.  A FRAMEWORK FOR GENERATING STOCHASTIC METEOROLOGICAL YEARS FOR RISK-CONSCIOUS DESIGN OF BUILDINGS , 2012 .

[6]  C. Reinhart,et al.  DAYLIGHT FACTOR SIMULATIONS – HOW CLOSE DO SIMULATION BEGINNERS ‘ REALLY ’ GET ? , 2009 .

[7]  Lorrie Faith Cranor,et al.  A Framework for Reasoning About the Human in the Loop , 2008, UPSEC.

[8]  Pieter de Wilde,et al.  Design analysis integration: supporting the selection of energy saving building components , 2002 .

[9]  David E. Bradley,et al.  Experiences with and interpretation of standard test methods of building energy analysis tools , 2004 .

[10]  Philip Haves,et al.  IMPACT OF MODELER DECISIONS ON SIMULATION RESULTS , 2014 .

[12]  Pamela Marie Berkeley.  Method for Testing and Classifying the Effect of the Modeler on Building Energy Simulation Results , 2013 .

[13]  Godfried Augenbroe,et al.  Uncertainty quantification of microclimate variables in building energy models , 2014 .

[14]  Robert W. Reeder,et al.  Improving user-interface dependability through mitigation of human error , 2005, Int. J. Hum. Comput. Stud..

[15]  J. Nielsen.  Usability inspection methods , 1994, CHI Conference Companion.

[16]  Wei Tian,et al.  A review of sensitivity analysis methods in building energy analysis , 2013 .

[17]  Christiaan J. J. Paredis,et al.  TOWARDS BETTER PREDICTION OF BUILDING PERFORMANCE: A WORKBENCH TO ANALYZE UNCERTAINTY IN BUILDING SIMULATION , 2013 .