The intermittent nature of renewable sources, uncertain demand load, and the location of microgrids cause challenges for the proper energy balance between supply and demand. To overcome the issues caused by such challenges and to be economically efficient while minimizing environmental impacts, microgrid design with renewable energy sources has recently attracted the attention of researchers. However, the sustainable microgrid design involving multiple types of demand areas (residential and industrial areas) and seasonal factors has not been explored by researchers. Therefore, this paper investigates the sustainable microgrid design problem with multiple types of demand areas and peer-to-peer energy trading involving seasonal factors and uncertainties to maximize total profit and to minimize environmental costs while satisfying demand. The problem is to determine optimal decisions on the number, location, and capacity of renewable distributed generation sources, energy flows, and seasonal electricity sales prices for the system that includes peer-to-peer trading. A fuzzy multi-objective programming model is developed to handle the uncertainties of demand load and capacity of renewable distributed generation sources. A genetic algorithm is applied to solve the proposed model. Results of computational experiments to assess its effectiveness and efficiency show that the proposed model increases total profit by over 12% and reduces environmental costs by almost 30% compared to the cases of no peer-to-peer energy trading and seasonal factors.