Sensitivity analysis – often executed simultaneously with Uncertainty Analysis (UA) – is a study which provides insights on the contribution of uncertain input parameters over the outputs of a model \cite{kurowicka2006uncertainty} . Uncertainty and sensitivity analysis generally follow two major objectives in the field of building energy performance. One group of studies evaluate the embedded uncertainty in early design evaluation to quantify the range of outputs \cite{Bucking_2013}, while others perform uncertainty and sensitivity analysis in order to validate and calibrate previous measurements\cite{Calleja_Rodr_guez_2013,westphal2005building}.
A vast variety of topics in building energy performance has been covered in recent studies. De Wit et al. \cite{de_Wit_2002} introduced a  methodology for decision makers to quantify uncertainties in the design  process. Such frameworks assist decision makers to deal with various design  possibilities that are derived from a series of sensitivity analyses. Other  studies have adopted comparable approaches to support decision making in  building energy assessment and the impacts of possible design solutions on the  energy performance \cite{Kim_2014,de_Wit_2002,Attia_2013a}. Eisenhower et al. \cite{Eisenhower_2012} introduced an uncertainty and  sensitivity decomposition model to study the effects of uncertain quantities on  a building’s energy performance. The proposed framework was aimed at evaluating  the impact of sizing sub-components of a model on the overall building energy  performance. Hopfe et al. \cite{Hopfe_2011} argued that including  uncertainty in physical parameters during the design process will ensure a higher quality a building’s energy performance. Moreover, it was noted that  assessing uncertainty in design parameters can provide designers with better  decision strategies. There has been a limited number of studies, which has focused on UA in building performance simulation with respect to uncertainty in weather parameters. Considering the complexity of propagating weather parameters through an energy model, uncertainty analysis has been either ignored or limited to a specific season \cite{Garcia_Sanchez_2014,Macdonald_2001,Corrado_2009}. Sun et al. \cite{Sun_2013} presented a quantification of uncertainty was addressed through the exploration of the urban heat island effect. Temperature, wind speed, wind pressure and solar radiation were considered as the most important climatic parameters related to exterior building boundary conditions that affect the overall energy performance of buildings.
Clearly, the outdoor climate is one of the main factors that  affect the overall performance of buildings. The outdoor climate is also affected by the Urban Heat Island effect (UHI) in large cities. Investigation on the  long-term hygrothermal performance of a residential building has demonstrated  the difference in design requirements for urban and rural contexts  \cite{paolini2016hygrothermal}.  According the recent studies, the Meanwhile, it has also shown that the impact of boundary conditions in smaller  scales of the built environment, may also affect the performance of the  building\cite{Karlsson_2006}. This has been investigated in a variety of studies related to the building envelope such as shading devices \cite{Tzempelikos_2007}, daylight \cite{Loutzenhiser_2007} and solar heat gains\cite{Lee_2013}.  The role of boundary conditions also discussed in the risk assessment of mold growth. The knowledge of the boundary conditions can be improved optimizing the modeling of some variables like weather parameters, such as the estimation of the solar irradiance incident on a facade\cite{Freitas_2015}.  Since the uncertainty in weather parameters is clarified in previous studies \cite{de_Wit_2002,sun2013uncertainty}  and according to literature the application of UA analysis in building outdoor boundary conditions merits to be considered.
Solar irradiance is among the weather parameters that affect different building energy performance parameters such as indoor air quality, visual comfort, building integrated photovoltaic system, etc. Building simulation tools compute the incident solar irradiance on building surfaces. Mostly these models use a set of observed global solar irradiance to compute diffuse solar irradiance. Solar models are developed based on data collected at different locations and different climatic conditions. However it is argued that solar models cannot be considered as generally valid models and the uncertainty in calculations based on these models is inevitable \cite{Prada_2014,pernigotto2015impact}.
The mentioned solar radiation models are applied to calculated diffuse and direct irradiation, however, the incident solar irradiance on the building surfaces (IH) is divided to three components. The direct normal solar irradiance (IDI), diffuse solar irradiation (IDIF) and ground reflected irradiation (IGR).
\begin{equation} I_{H}=I_{\text{DI}}+I_{\text{DIF}}+I_{\text{GR}}\nonumber \\ \end{equation}
According to literature, another source of uncertainty in IH arises from the reflected irradiations. However, the model of calculations for these parameters (in the area of building energy analysis), is a simplified version of reality (Fig. 2). Recent studies tried to expand the mentioned equations in order to consider more dominant parameters, which affect the amount of incident solar irradiance on the building surface. Lou et al \cite{Lou_2016} evaluated the solar irradiation on obstructed building façade. To evaluate the reflectance of adjacent buildings, they proposed an additional value to equation (1):
\begin{equation} I_{H}=I_{\text{DI}}+I_{\text{DIF}}+I_{\text{BR}}+I_{\text{GR}}\nonumber \\ \end{equation}
Where the IBR is related to reflected solar irradiation from adjacent buildings.
Proposing a more accurate model of solar radiation on the building façade calls for considering more detailed variables. This issue caused other studies to focus on the role of vegetation on the uncertainty of incident solar irradiation within the street canyon \cite{Fogl_2016,Krayenhoff_2013,Bourbia_2004}. Cascone et al \cite{Cascone_2011}investigated the impact of different shapes of vegetation in different orientations and seasons on the shading factor. In their investigations, they found about 24.1% difference of shading factor in the case of various foliage shapes on the south façade and the same conditions for the east façade showed the 23.9 % difference of shading factor.