RESEARCH / DEVELOPMENT / TECHNOLOGY PIPELINE TECHNOLOGY JOURNAL 51 in a region characterized by: (a) great potential for offshore development in the near future, and (b) high seismicity and consequent earthquake-related geohazards. Although the new tool requires further improvement, the preliminary results demonstrate its capability to handle, analyze and manage all the available spatial data that are directly or indirectly linked with the earthquake-related geohazards and to support the geoscientists and engineers to quan- tify the geohazards and the relevant risks, and to make a prompt and clear distinction between the actual critical ar- eas that a pipeline cannot cross and the non-critical areas that a pipeline can safely cross. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Apel, E., Bürgmann, R., Serpelloni, E., 2007. Rigid Block Motion, Interseismic Strain, and Backarc Deformation in the Aegean, in: AGU Fall Meeting Abstracts. ESRI (Environmental Systems Resource Institute), 2016. ArcGIS Desktop: Release 10.4. Redlands CA. Gennesseaux, M., Winnock, E., 1993. Thickness of Mediterranean plio-quaternary sediment [WWW Document]. Charts Div. Head Dep. Navig. Oceanogr. Russ. under Auth. Intergov. Oce- anogr. Comm. UNESCO. URL https://www.ngdc.noaa.gov/mgg/ibcm/ibcmsedt.html. Lagaros, N.D., Tsompanakis, Y., 2006. Intelligent computational paradigms in earthquake engineering, Idea Publi. ed. Makrakis, N., Psarropoulos, P., Chatzidakis, D., Tsompanakis, Y., 2020. Route Optimization of Offshore Lifelines Taking Into Account Potential Earthquake-Related Geohazards. Front. Built Environ. 6, 1–16. https://doi.org/10.3389/fbuil.2020.00112 Psarropoulos, P.N., Antoniou, A.A., Tsompanakis, Y., 2019. Offshore earthquake-related geoha- zards and route optimization of offshore pipelines, risers, and cables, in: Pipeline Technology Conference, Berlin. Psarropoulos, P.N., Makrakis, N., Boutikas, K., Papathoma, M., 2021. Artificial Neural Networks in the Design of Offshore Pipelines against Geohazards: Pipeline Technology Conference, Berlin. Randolph, M., Gourvenec, S., 2017. Offshore geotechnical engineering, Offshore Geotechnical Engineering. https://doi.org/10.1201/9781315272474 Simulia, 2014. Abaqus 6.14 analysis user’s manual. The MathWorks Inc., 2015. Matlab Release 2015a [WWW Document]. Massachusetts, United States. Trimintziou, M.S., Sakellariou, M.G., Psarropoulos, P.N., 2015. Designing Offshore Pipelines Facing the Geohazard of Active Seismic Faults Designing Offshore Pipelines Facing the Geoha- zard of Active Seismic Faults. Int. J. Civil, Environ. Struct. Constr. Archit. Eng. Tsompanakis, Y., Lagaros, N.D., Stavroulakis, G.E., 2008. Soft computing techniques in parame- ter identification and probabilistic seismic analysis of structures. Adv. Eng. Softw. https://doi. org/10.1016/j.advengsoft.2007.06.004 Authors Prodromos Psarropoulos National Technical University of Athens Structural & Geotechnical Engineer prod@central.ntua.gr Nikolaos Makrakis Independent Consultant Surveying & Earthquake Engineer makr.nickos@gmail.com Figure 7: The final pipeline route that has been proposed by the decision-sup- port tool. Table 1: Errors of the ANNs for U1 and ε. of these neural networks to predict random data belonging to the range of the preliminary FE analyses was tested. Ta- ble 1 shows the errors in the prediction of U1 and ε, where it becomes evident that ANNs can predict U1 very efficiently, with maximum errors less than 1%, while the prediction of ε has a less reliable performance. The results were compared with the corresponding data of the neural network for method testing. The prediction error for the axial strain of the pipe was smaller than 0.5%. In conclusion, the objectives were met in a sufficient degree. It was proved that ANNs can predict the behaviour of sub- sea pipelines, which are subjected to the displacements of active faults’ rapture, provided that they are based on datasets that are extracted from accurate FE simulations. 6. CONCLUSIONS The current study focuses on route optimization of offshore pipelines taking into consideration the poten- tial crossing of extensive submarine areas, facing the geohazard of active fault rupture. Combining the capabil- ities of ArcGIS platform with the FE software ABAQUS, a smart decision-support tool had been developed in the past by the authors and their colleagues to facilitate: (a) the qualitative and quantitative assessment of the major earthquake-related geohazards along a possible lifeline routing, (b) the quantitative assessment of their potential impact on the lifeline, and (c) the selection of the optimum pipeline route. In the current study the standard version of the smart tool has been improved with the application of AI. The tool has been applied in a characteristic case study