Sensitivity Analysis of Hygrothermal Performance of Cross-Laminated Timber Wall Assemblies

Cross-laminated timber (CLT) panels are increasingly being used in building enclosures due to their good structural and fire safety performance. However, prolonged exposure to moisture during construction and in service are durability concerns for most wood products, including CLT. The wetting and drying behavior of CLT wall assemblies can be studied by hygrothermal simulations in which a deterministic approach is normally used. However, in reality, there are always uncertainties in input parameters—such as material properties, environmental loads, and design variables—that may lead to discrepancies between simulation results and actual performance. The hygrothermal performance of 16 CLT wall assemblies with various design configurations was tested in a building envelope test facility, and discrepancies between simulations and measurements were observed. This paper further investigates the discrepancies between simulations and measurements of a CLT wall assembly with two different types of water-resistive barriers (WRBs) that were caused by the uncertainties of input parameters using sensitivity analyses. Simulation results obtained from DELPHIN and WUFI Pro simulation programs are compared with measurements for validation. The influential factors—including material properties, rain loads, and cladding ventilation rates—are studied using a one-factor-at-a-time method under different environmental loads. The examined parameters are assigned with two extreme values based on their uncertainties. The root mean square difference of CLT moisture content between the cases with the two extreme values is calculated to evaluate the importance of each parameter. The simulation results show that the influence of the moisture storage function is more significant than the moisture transport properties (i.e., vapor resistance factor and moisture diffusivity) and that the wall assembly with a vapor-permeable WRB is more sensitive to the variations in the rain deposition factor and cladding ventilation rate than the wall with a non-vapor-permeable WRB.