At the early development phase of the safety-critical embedded system, we examine the ability of hardware-software co-design to meet the target reliability/availability. Researches have reported interaction between software and hardware plays a major role in system operations. They have categorised the interaction failures as following: hardware-related software and software-related hardware. The existing reliability/availability models either have not considered all types of interaction failures or provided qualitative analysis rather than quantitative reliability/availability. They also have not consider the operational profile of the system. We propose a quantitative reliability/availability prediction model considering entire spectrum of interaction failures and also take into account system operational profile to achieve more precision over prediction. At first, we identify system functionalities from the system requirements specification (SysRS) document. Then, we map these functionalities to functionally equivalent conceptual design components. Subsequently, we configure the system based on the flow of functional execution. Then, we perform random sampling of components operation scenarios to test system functionalities, using digital simulation platform. Finally, we predict system reliability/availability based on simulation result. In this paper, we contribute a consolidate reliability/availability prediction model that addresses interaction failures (hardware-related software and software-related hardware failures) along with hardware specific or software specific failures. We also apply the proposed model to a case study and compare the obtained result with the continuous-time Markov chain based model for validation.