With the advent of software technologies, over a period of time, the Industrial Control Systems (ICSs) have grown exponentially. Whereas, almost all ICSs comprise Human Machine Interfaces (HMIs), which are the key component for monitoring and controlling complex industrial systems. For decades, traditional HMIs with simple User Interfaces (UIs) remained operational to minimize the complexities and resulting operational costs. However, due to the emergence of smartphone technologies, the perception about user interfaces has been transformed significantly and users now demand same sort of experience with industrial HMIs, as well. There are few industrial solutions, like, ICONICS GraphWorX to support the development of mobile HMI screens. However, such proprietary solutions are quite expensive. Furthermore, the underlying development approaches and source codes are not accessible in public domain. On the other hand, the state-of-the-art approaches for the development of native mobile HMI screens are hard to find in the literature. Consequently, there is dire need of a cost-effective, easy to use, open source framework for the development of native mobile HMI screens. In order to achieve this goal, here we propose, a Model-driven Mobile HMI Framework (MMHF). MMHF comprises, a Unified Modeling Language (UML) Profile for Mobile HMI (UMLPMH) for modeling of HMI screens. MMHF also includes, an open source transformation engine and a Model Driven Mobile-based HMI Code Generator (MDMHCG) to automatically transform UMLPMH models into target native mobile HMI implementations. Consequently, MMHF enables simpler way to design the HMI screens using UMLPMH and generates native Mobile HMI Screen implementations automatically using MDMHCG. The empirical evidence of MMHF is demonstrated through three (3) benchmark case studies, which prove that the MMHF is a feasible, cost effective and scalable solution to develop native HMI screens for wide-ranging ICSs. INDEX TERMS Human machine interface, mobile HMI, model driven engineering, unified modeling language, industrial control system, industry automation, Internet of Things (IoT).