Energy is an essential input for social and economic development throughout the world. Ensuring sufficient quantity and satisfactory quality in regional energy systems requires the formulation of a model to allocate energy resources and to properly develop renewable energy sources. In addition, the contradictions and interactions of economic development and environmental sustainability are significant concerns in energy system management. Accordingly, effective planning of regional energy systems is a priority for social, economic, and sustainable environmental development. In this research, an inexact chance-constrained mixed-integer programming (ICMI) model was developed to manage a regional energy system and greenhouse-gas (GHG) emissions under uncertainties in Yukon. The local policy interests in using renewable energy and managing climate change for regional energy systems were taken into consideration. The proposed model employs mixed-integer linear programming to determine the optimal capacity expansion of generating facilities to meet demand, interval-parameter linear programming to tackle the system uncertainties expressed as interval numbers, and chance-constrained programming under a series of credibility levels to evaluate the system reliability under different availabilities of renewable energy. In addition, an interval credibility-constrained integer programming (ICIP) was proposed for community-level regional energy planning. Through integrating interval programming, credibility-constrained programming and mixed-integer programming, the ICIP model can address uncertainties expressed as fuzzy set and interval numbers. The model incorporates the concept of credibility to account for the decision makers’ level of confidence in energy demand. The developed ICIP model was applied to a hypothetical regional energy management problem. Tradeoffs between system costs and risk represented as constraint violations could be tackled. Decision alternatives for the energy supply scheme and desired capacity expansion with minimal system cost and maximum system security were provided. The developed models can provide solutions of energy demand-supply and capacity expansion schemes. Uncertainties expressed as intervals, probability and possibility distribution in the system can be effectively addressed. The results indicate that the developed models are applicable to supporting decision making in energy systems and environmental management.