This paper examines the impact of location and load shape selection on microgrid optimal design. 96 unique combinations of location and load shape are considered to provide a broader scope than any previous work in microgrid design sensitivity analysis. In addition, the level of autonomy from the macrogrid is considered as a tunable parameter in the optimization. A generic system is considered consisting of photovoltaics, wind turbine, microturbines, electric and natural gas boilers, thermal storage, and a battery bank. The microgrid is grid-connected and designed to supply both heat and power. A mixed integer linear program is used to minimize the expected cost of energy supply over a 20 year horizon. Trends in the design results are discussed and important input parameters that depend on the location and load shape are identified. Finally, a procedure to quantify these trends and predict optimal design results in new locations is proposed. (C) 2015 Elsevier Ltd. All rights reserved.