The emergence of 5G and social networks, along with the proliferation of end user smart devices and enhanced applications, has created and exponentially increased the interest in the broader area of mobile cyber-physical social systems. Typically, these systems present competitive environments that induce constraints, while users evolve in a physical, digital or virtual space with others, where their decisions and behaviors become interdependent. Realizations of such systems include 5G wireless communication systems, Internet of Things (IoT) infrastructures, Mobile Edge Computing (MEC), cyber-physical systems, and the intersection of these technologies with other disciplines, such as smart cities. REALISM aims at introducing a paradigm shift towards providing satisfaction, realism and efficiency in such interdependent networking and resource sharing environments. In particular, towards utilizing the available resources more efficiently, we argue that better energy-efficiency is achieved by targeting satisfactory QoS levels only (through the introduction of satisfaction equilibrium), rather than QoS maximization which is generally energy costly, Moreover, we provide a real-life spirit and more pragmatic approach into the resource management process and deal with the challenges in the decision-making due to user risk preferences, that traditional models fail to capture. Despite the generality and holistic nature of the introduced framework, in this proposal the focus is placed on emerging communication paradigms, e.g., 5G wireless access technologies, and MEC environments, where multiple resources of different nature and properties need to be optimally allocated.