Renewable energy penetration has been steadily increasing in many countries. Governments have created policies that encourage investment and announced long-term objectives for the inclusion of renewable sources. However, specific installation plans for distribution systems with the presence of distributed generators of renewable origin have not been discussed yet in Colombia. This study presents a two-stage stochastic programming solution based on scenarios to solve the problem of distributed generation allocation, computationally tested on real a 34-bus distribution network located in Colombia. The procedure for generating scenarios, based on observed meteorological data, is performed using K-means on measured data. Additionally, a sensitivity analysis is performed for the evaluation of the optimal solution based on the investment budget.