MODELO DE RED NEURONAL ARTIFICIAL APLICADO A LA VALORACIÓN DE ACTUACIONES DE ADECUACIÓN DE HABITABILIDAD EN ACTIVOS INMOBILIARIOS

Abstract

Nowadays, one of the main sector with the greatest demand in the real estate sector is the management of real estate assets. In this regard, the adequacy of the existing real estate stock is a broad field of work that ranges from technicians to operators. The habitability adaptation process of these assets is an activity with a large consumption of temporary resources and involves a complex organization of tasks that makes the process of prioritizing the actions difficult. Because of this circumstance, there are techniques of artificial intelligence, such as artificial neural networks, which can facilitate the analysis and assessment process. In this work, a multilayer perceptron is proposed for one of the most numerous types of real estate assets: dwellings. For this purpose, a total of 200 different dwellings are analysed, which are used for the training and validation of the model. The results allow to predict with a high degree of adjustment the priority of action and the estimated amount of the habitability adaptation works.