Most of electricity interruptions are produced in distribution systems, and one of the solutions to increase levels of reliability in this area are urban microgrids. This work seeks to design urban community microgrids through a stochastic optimization model, finding their PV and storage systems adoption. Investment, operation and value of lost load, for multiple community sizes to study the benefits that community association can bring. Hundreds of urban community microgrids were designed, because energy needs of households differ among them, therefore all designs are different. As more households are grouped in urban community microgrids, investment and operation costs of design per client decrease. However, larger communities mean that the microgrid has more internal network extension (to connect the members of the microgrid), which translates in a higher probability of failure than an individual household, hence there is a trade-off between benefits in investment and operation costs per household and failure rates. For this reason, value of lost load is included and grows as more households are grouped as communities and thus, the optimal number of members of a microgrid can be determined. In this work, microgrids are designed to operate under four different business models, obtaining the less expensive annual costs when the microgrid can use its assets to directly meet the energy needs of a given community. Additional cost of upgrading into an urban community microgrid is found for all community sizes, being optimal for communities of 25 or 50 households depending on the business model and value of lost load. Results show that implementing urban microgrids is feasible at current prices, and for households it would represent an additional cost (above current tariffs) of 11 % on average per year for communities, while 52 % if the microgrid has only one member.