Sustainability and energy consumption awareness led industrial sector to reduce energy consumption. This reduction is regarded as a solution to reduce greenhouse gas emissions. Moreover, international regulations about maintenance activities involve hazardous energy-any electrical, mechanical, nuclear or other energies that can harm personnel- as a rising threat. Thus, energy audits and diagnosis of existing manufacturing systems are crucial to achieve energy efficiency. Future manufacturing paradigms as reconfigurable manufacturing system (RMS) have shown high responsiveness to cope with new challenges such as sustainability. This paper proposes a sustainable RMS design through process plan generation. The approach is developed to generate a process plan while diagnosing energy flow and assigning preventive maintenance activities related to reliability reduction in system components. More specifically, a mixed-integer non-linear program is proposed, then solved using a bi-level decomposition approach. The lower-level considers process plan generation following parts requirements and guided by energy loss as an objective. Afterwards, the upper-level diagnoses the reliability of the lower-level selected machines and tools. Moreover, it checks if preventive maintenance is required due to the level of hazardous energy and maintenance plan. The approach applicability is validated through an illustrative example.