**4. Conclusions**

The comparison of ANFIS estimation results against SWERA, and PVGIS values for specific validation sites shows a reasonable approximation to both benchmark validation data sets. Both data sets are produced using remote sensed data in combination with NWP and validated by local meteorological stations. Seeing the correlation plots of the parameters it could be demonstrated that the ANFIS technique produces a robust model which follows very well the patterns of the training data. On the other hand it can be noted that with the proposed methodology, good results could be achieved in-spite of limited input data. Finally, it can be said that this methodology is creative and unique in its approach on GIS-based solar assessment as well its application to the case study in Ethiopia.

The main contributions of this work are:


The presented case study is a simplified training procedure for ANFIS estimation of radiation, even though this procedure is reproducible at any scale and able to contain more parameters in order to improve its performance.

**Author Contributions:** Conceptualization, B.A.W. and J.A.D.-N.; methodology, B.A.W.; software, B.A.W.; validation, B.A.W. and H.B.; writing—original draft preparation, B.A.W. and H.B.; writing—review and editing, J.A.D.-N.; supervision, J.A.D.-N. All authors have read and agree to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.
