The advanced computational power, together with the huge availability of data, has led to an unstoppable development of Artificial Intelligence, whose applications extend to several domains, such as healthcare, finance and computing science among others. The employment of Deep Learning and Machine Learning models involves accepting a tough challenge: deriving benefits while minimizing risks due to the black-box nature of Artificial Intelligence based-systems. To ensure trustworthiness, the basic key principles (Sustainability, Accuracy, Fairness, Explainability) have to be fulfilled. Recently, a sound approach for the assessment of the reliability of highly complex models was introduced by means of the formalisation of specific metrics, each one referred to the evaluation of the single principles. In this paper, we combine the SAFE principles into a unified indicator addressed to set the trustworthiness degree of Machine and Deep Learning models, providing a tool for ranking the models in terms of safety.