The heterogeneous deployment of devices that originate from the 5G New Radio networks is of utmost importance. The advantages of such technology offer connectivity, low latency, and energy efficiency. There are three bands in the New Radio, namely the lowband, the mid-band, and the mmWave band. In this paper, we address the mid-band for Outdoor-to-Indoor communications. The primary objective of this paper is to suggest an explainable Artificial Intelligence model, which will assist in identifying the impact of dataset features on the prediction of the model. We identified a publically available dataset, which included specific features of measurements, such as the power, total powers as well as coordinates and delay of communication. To this end, we attempted to predict the delay using a Random Forest model and we utilised the SHapley Additive exPlanation framework to explain the model. This provides helpful insights regarding which features contribute the most to the prediction as well as their impact and correlation. Moreover, we show that with the performance of the model, the explainability of Artificial Intelligence is very useful. Finally, we show the contribution of each feature in multiple plots and show that this explainability model can be useful to datasets that need to indicate the impact of features to the prediction.