Abstract
This article addresses the problem of locating temporary shelters in small Colombian cities through the construction of a stochastic linear programming model that considers different scenarios where the affected population is a factor that generates uncertainty in the location of shelters and the respective flows in the humanitarian supply chain. The model is validated through the construction of a case study referring to the floods presented in a city in the center of the department of Valle del Cauca, which constantly causes the need to relocate the affected population, seeking to reduce attention times to the victims while considering restrictions associated with the budget available for emergency care. The generated proposal becomes a benchmark for the efficient management of shelters in similar disasters in small Colombian cities.