Uncertainty and Sensitivity Analyses of Switchable Slat Insulated Shades for Daylighting Performance based on Illuminance Thresholds and Objective Weights

To balanced multi-criteria's daylighting performance in indoor spaces, several dynamic metrics have been proposed, but so far there is no convention on which daylight metrics thresholds are preferred and which objective weights are given priority in optimization of daylighting under certain climate. This study explores uncertainty and sensitivity analyses to identify and prioritize the most influencing critical objective weights and appropriate useful daylighting illuminance (UDI) threshold for designing glazed system with Insulated Glazed Units that include Parallel Slat Transparent Insulation Materials (IGUs/PS-TIMs) for office buildings in sub-tropics areas for achieving best level of daylighting availability with enhanced thermal and visual comfort. To ascertain which objectives and threshold have significantly impact the optimized IGUs/PS-TIMs parameters, the study performed a framework of three statistical methods-based multi-objective optimization (MOO) of Objective’s weights (OWs), Sensitivity analysis (SA) and uncertainty analysis (UA), were used. The framework comprises multi-optimization stages based on two main steps. In the first step, MOO based-multi statistical analysis was run for ranking UDI threshold control setpoint (UDI-TCS) and of critical objective weights (PC-OWs), meanwhile the second step, MOO based-optimal UDI-TCS and PC-OWs was run for final ranking solutions. The results showed that if considering the priority of PC-OWs in ranking final solutions based best UDI thresholds, the different schemes can be obtained, which is of great significance to the early design stage of buildings. Overall, the results showed that when considering PC-OWs in ranking final UDI thresholds, it can be clearly noticed that the UDI, ASE, and QV respectively, are the most PC-OWs, and UDI500-2000lux is the optimal threshold because it has significantly improved the UDI and ASE in all optimal cases compared to optimal cases resulted from UDI500-1000lux/50%. the total average percentage of UDI between 22.63% to 27.80%, meanwhile, the total average percentage of ASE improved between 2.14%. to 11.52%. For QV, the optimal cases result from UDI500-1000lux/50% threshold working better by slightly improving QV by 1.16% and 10.56%. Notwithstanding, these two pairs of UDI thresholds with given priority to UDI, ASE and QV as PC-OWs are suggested as the most appropriate conflicted objectives for optimizing daylight under sub-tropical climate conditions.