Toward AI-Assisted Conceptual Design of Subsea Inline Structures: Classification and Strain Behaviour of Inline Components during S-Lay Installation

Abstract

S-lay installation remains the dominant method for offshore pipeline deployment, with inline components — valves, flanges, Inline Tee (ILT) assemblies, and Pipeline End Terminations (PLETs) — increasingly integrated into the continuous pipeline string. As these structures traverse the stinger, the displacement-controlled overbend condition induces strains in the adjacent pipeline that exceed those in plain pipe, a complex and highly localized design condition unique to such subsea assemblies. For a given piping layout, many structural configurations are possible, and identifying the one that minimizes installation strain is a non-trivial search problem that fast-track project schedules rarely allow engineers to conduct systematically; designs are instead adapted from past projects using individual pattern recognition rather than a structured body of design knowledge. This paper is the first in a series developing an AI-assisted conceptual engineering framework for subsea inline structures, using the S-Lay ILT design problem as its case study. It introduces a dual classification system — a physical taxonomy (IW/EA) describing how components connect to the pipeline, and a mechanical taxonomy (Types A, B, C) describing their strain-amplification behaviour — and characterizes each behaviour type through parametric finite element analyses incorporating stinger–pipeline contact modelling. Results identify offset depth and component length as the dominant strain drivers for elevation- and stiffness-type components respectively, quantify the DNV-compliant range for taper transitions, and show that very long components can paradoxically reduce peak strain through dual-roller contact. Together, these findings constitute the design intuition — Steps 1 and 2 of the proposed seven-step framework — needed to support AI-assisted conceptual design of subsea structural assemblies

Introduction

This paper presents the initial stage of a long-term framework for AI-assisted conceptual design of subsea inline structures (such as Inline Tees (ILTs) and Pipeline End Terminations (PLETs)) used in offshore pipeline installation. These structures consist of multiple interacting components whose combined behavior during installation is difficult to predict. Because engineers often rely on experience and previous designs rather than systematic optimization, the paper proposes creating a structured, component-level knowledge base that AI can use to generate and evaluate design concepts before detailed finite element analysis.

The study focuses on the S-Lay pipeline installation method, where pipelines pass over a curved stinger supported by rollers. During this overbend stage, pipelines experience significant bending, making it a critical region for strain control. Unlike J-Lay or Reel-Lay methods, S-Lay introduces discrete roller contact, which can produce localized strain concentrations, particularly when large inline structures pass over the stinger.

The paper explains the role of the S-Lay vessel and stinger, where articulated roller boxes guide the pipeline while controlling curvature and minimizing damage. Previous research has shown that pipeline stresses in the overbend depend on factors such as stinger geometry, roller spacing, pipe tension, vessel motion, and environmental loading, highlighting the need for accurate numerical analysis and optimized structural design.

Special attention is given to inline structures such as ILTs and PLETs, which are integrated into the pipeline during installation. These structures introduce localized stiffness changes, increased weight, and geometric discontinuities. As they pass over stinger rollers, they experience combined axial, bending, and contact loads that can create stress concentrations and increased pipeline strain. Components such as valves, tees, flanges, and protective base structures further complicate the structural response by locally lifting the pipeline above the rollers and increasing curvature.

The paper introduces a seven-step AI-assisted engineering framework, with this work addressing the first two stages:

1. Component Classification – identifying and categorizing fundamental subsea structural components.
2. Design Intuition – understanding how individual components influence strain amplification during installation.

Future stages will build a validated database of component behaviors that AI can use to assemble and assess complete inline structures, enabling rapid concept generation and informed engineering decisions before detailed simulation.

Conclusion

This paper introduced a dual-classification framework — a physical IW/EA taxonomy describing how inline hardware connects to the pipeline, and a mechanical Type A/B/C taxonomy describing the strain-amplification mechanism each connection produces — that bridges the hardware design domain (DNV-ST-F101, ASME VIII) and the S-Lay installation analysis domain. Through five progressive series of Abaqus parametric finite element analyses, it characterized the isolated and combined behaviour of Types A1, B1, and C1: – Type A1 (stiffness-dominant): taper slope has negligible effect within the DNV-compliant range (1:4–1:16, <5% difference) but produces a step-change strain penalty beyond it; thick-section length is the dominant parameter and becomes critical beyond ~10D; inter-component spacing has negligible effect; very long components can enter a dual-roller relief regime. – Type B1 (elevation-dominant): offset depth V is the dominant, near-linearly-acting parameter; deep-section length and taper slope are largely insensitive in the normal range; very long shrouds (?25D) trigger the same dual-roller relief mechanism seen in Type A. – Type C1 (combined): always exceeds both isolated Type A and Type B responses; the location of a stiffness discontinuity within an elevated zone is the critical design variable, with placement in the catenary-side peak-strain region producing an 18–22% penalty over placement elsewhere. Together, these findings constitute the quantitative design intuition — Steps 1 and 2 of the seven-step AI-assisted conceptual engineering programme proposed in Section III — needed to support informed, code-aware conceptual design of subsea inline structural assemblies before detailed finite element analysis is undertaken

References

[1] Y. Bai and Q. Bai, Subsea Pipeline Design, Analysis and Installation. Oxford: Gulf Professional Publishing, 2014, ch. 33. [2] A. Palmer and R. King, Subsea Pipeline Engineering, 2nd ed. Tulsa: PennWell, 2008. [3] D. Perinet and I. Frazer, “Strain Criteria for Deep Water Pipe Laying Operations,” Offshore Technology Conference, OTC 19329, Houston, Texas, May 2008. DOI: 10.4043/19329-MS. [4] R. O’Grady and A. Harte, “Localised assessment of pipeline integrity during ultra-deep S-lay installation,” Ocean Engineering, vol. 68, pp. 27–37, 2013. [5] DNV-ST-F101, Submarine Pipeline Systems. Det Norske Veritas, October 2017. [6] K. Huang, A. Ji, and E. Uribe, “Deepwater In-Line SLED Installation Methods and its Application to Frade Project,” Offshore Technology Conference, OTC 19805, Houston, Texas, May 2009. [7] J. K. Antani, W. T. Dick, D. Balch, and T. van der Leij, “Design, Fabrication and Installation of the Neptune Export Lateral PLETs,” Offshore Technology Conference, OTC 19688, Houston, Texas, May 2008. [8] H. Mota, “Development of In-Line PLETs for TUPI Pilot Project: Challenges and Lessons Learned,” Offshore Technology Conference Brasil, OTC 22182, Rio de Janeiro, Brazil, October 2011. [9] Xie, P. et al., Overbend monitoring during deepwater S Lay, International Journal of Offshore and Polar Engineering, 2017 [10] Xie, P. et al, Real-time strain monitoring in overbend, Proceedings of the Twenty-fifth International Ocean and Polar Engineering Conference, 2015 [11] Ivi?, S. – Sensitivity analysis of S Lay configuration, International Journal of Applied Engineering Research, 2015 [12] Xu, P. & Gong, S. (2020) – Dynamic response of S Lay pipelines under waves, Journal of Offshore Mechanics and Arctic Engineering, 2020 [13] S. Kyriakides and E. Corona, Mechanics of Offshore Pipelines, Vol. 1: Buckling and Collapse. Oxford: Elsevier, 2007. ISBN 978-0-08-046732-0. [14] Y. Liu and S. Kyriakides, “Effect of Geometric and Material Discontinuities on the Reeling of Pipelines,” Applied Ocean Research, vol. 65, pp. 238–250, 2017. DOI: 10.1016/j.apor.2017.04.004. [15] R. Bruschi, F. Torselletti, E. Vitali, B. Marchionni, P. Ragni, and G. Skaug, “Controlling Fatigue Damage during Deepwater Installation of Inline Components,” ASCE Journal of Pipeline Systems Engineering and Practice, vol. 8, no. 3, pp. 04017003, Aug. 2017. [16] G. Kaliyaperumal, O. Fang, and D. Sealock, “A Challenging Trunkline ILT Design Driven by S-Lay Installation,” Offshore Technology Conference, OTC 24411, Houston, Texas, May 2013. [17] J. J. Jansen, “Partially Displacement Controlled Conditions in the Overbend During S-Lay Pipeline Installation,” M.Sc. thesis, Delft University of Technology, Delft, Netherlands, May 2018. [18] M. Liu, H. Zhou, B. Wang, Y.-Y. Wang, J. Bergman, B. Ayton, M. Stephens, T. Weeks, and J. Gianetto, “Strain-Based Design and Assessment in Critical Areas of Pipeline Systems with Realistic Anomalies,” Final Report, Contract No. DTPH56-14-H-00003, U.S. Department of Transportation PHMSA, August 2017. [19] H. D. Yun, R. R. Peek, P. R. Paslay, and F. F. Kopp, “Loading History Effects for Deep-Water S-Lay of Pipelines,” Journal of Offshore Mechanics and Arctic Engineering, vol.?126, no.?2, pp.?156–163, May 2004. DOI:?10.1115/1.1710871. [20] S. Ai, L. Sun, L. Tao, and S. C. Yim, “Modeling and Simulation of Deepwater Pipeline S-Lay With Coupled Dynamic Positioning,” Journal of Offshore Mechanics and Arctic Engineering, vol.?140, no.?5, p.?051704, Oct. 2018. DOI:?10.1115/1.4040049. [21] Orcina Ltd, “Orcaflex for Pipelay”, https://www.orcina.com/wp-content/uploads/OrcaFlex-for-pipelay.pdf, accessed June 2026

Copyright

Copyright © 2026 Sreekanth Manakkattil Sivaraman, Jagannatha Venakataramana Reddy. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.