Panama Metro Line 3

Engagement scope (across both phases)

• Phase 1 (2022) — full regional hazard model for the Line 3 alignment, covering the elevated viaduct sections, the TBM tunnel section and the cut-and-cover and NATM transitions
• Site-response analyses at multiple kilometre points along the alignment, accounting for the differing soil columns from Panama City to West Panama
• Independent peer review by Jacobs — the reviewer engaged by the HPH consortium for the line's engineering due diligence
• Phase 2 (2023) — pounding analysis at the interface between the special and standard viaduct sections (V8-6) where the structural typology changes
• Non-linear time-history simulation of the impacts between adjacent viaduct decks, using a Hertz-damped contact model
• Damper design at the special-viaduct interface to mitigate the pounding forces and protect bearings, deck edges and the adjacent station structures
• The same regional hazard source model from Phase 1 was later reused for the seismic engineering of the Panama Canal Fourth Bridge — every Panama project we have signed since reinforces it

Why this engagement mattered

Line 3 of the Panama Metro is the corridor that connects Panama City — at the eastern terminus of Albrook and the canal — to the rapidly expanding suburbs of West Panama on the far side. It is a mostly-elevated metro line with a tunnel section running under the mountain ridge and a cross-canal segment that shares its alignment with the new Fourth Bridge. When it opens it will move tens of thousands of passengers a day across the canal — disruption to it is a regional disruption, not a local one.

Three things make Line 3 a different problem from any other metro Dynamis has worked on. First, the alignment crosses three structural typologies in series: long elevated viaducts on piles, a TBM-bored tunnel section, and the cable-stayed canal crossing. Each typology sits on its own soil column and has its own dynamic signature — but they are linked by the same train, the same operator, and the same single rupture. Second, the active Pedro Miguel and Limón faults run through the canal corridor, contributing near-source crustal motion to a regional hazard otherwise dominated by the Cocos and Nazca subduction interfaces. Third, the line crosses the canal — there is no fallback corridor at the same scale.

What was actually delivered

Dynamis was retained by SENER Ingeniería y Sistemas, the Spanish engineering firm leading the project's design package, working under the contractor consortium HPH (Hyundai E&C – POSCO – Hanjin Heavy Industries). The engagement ran in two distinct phases, each with its own deliverable family.

Phase 1 (2022 → early 2023) built the seismic foundation of the entire line. Dynamis produced the full regional hazard model for the alignment — capable of feeding spectra and ground motions to the elevated viaduct sections, the TBM-bored tunnel, the cut-and-cover transitions and the NATM (New Austrian Tunnelling Method) sections in equal measure. Site-response analyses were run at multiple kilometre points along the alignment to capture the way the soil column changes between Panama City and the West Panama hills. The complete deliverable package was independently peer-reviewed by Jacobs, the international reviewer engaged by the HPH consortium for the line's engineering due diligence. That same regional source model would later be reused as the starting point for the hazard work on the Panama Canal Fourth Bridge — the Line 3 hazard study is the foundation on which Dynamis's broader Panama portfolio now sits.

Phase 2 (2023) went deep on a single, specialised problem: pounding at the V8-6 interface, where the line's standard elevated viaduct meets a special viaduct typology. Pounding is the seismic phenomenon that brings two independent structures into collision through their expansion joint when ground motion drives them out of phase — historically responsible for some of the most spectacular bridge-deck unseatings (Loma Prieta 1989, Kobe 1995). Dynamis ran the non-linear time-history simulations of the impact response with a Hertz-damped contact model, quantified the forces at the interface and on the adjacent pier and station, and designed the dampers needed to absorb the pounding energy and protect bearings, deck edges and adjacent structures. The deliverable went directly into SENER's design package for that specific interface.

What this means for portfolio risk

For a (re)insurer or a multilateral lender carrying transport-infrastructure exposure across Latin America, three observations from this engagement matter more than any single number:

1. One line, three typologies, one rupture. Long-elevated viaduct, TBM tunnel and cross-canal cable-stayed segments have very different damage curves under the same ground motion. A vulnerability function calibrated on "metro" as a single asset class will mis-distribute losses across the alignment — and badly miss the points where typology transitions concentrate damage.
2. Pounding is a known damage driver that generic models do not see. The standard-to-special viaduct interface is a textbook pounding-vulnerable detail. Without a project-specific analysis and a damper design, this single point of the alignment carries a disproportionate share of the line's PD/BI loss potential — and a generic country-level fragility cannot resolve it.
3. A peer-reviewed regional hazard model is portfolio-level evidence, not project-level paperwork. The Line 3 hazard study has been re-used by Dynamis for every subsequent Panama engagement, including the Fourth Bridge. For an underwriter pricing Panama exposure, the same source-model framework now sits behind multiple critical-infrastructure assets — that consistency is itself a line item in the diligence.

From engagement to portfolio indicators

Every project of this calibre — engineered to perform across three structural typologies on a single alignment, peer-reviewed by an international consultancy, with a dedicated specialist study on pounding and damper design — becomes evidence in the engine that powers Xpectral. The regional hazard model, the site-response analyses, the non-linear contact simulations, the damper specifications: none of them is an isolated deliverable. Together they calibrate the hazard model and, more importantly for our roadmap, they inform the engineering-grade fragility curves we are now building for the asset classes (re)insurers and multilateral lenders cannot afford to misprice.

Fragility curves today are built on simplified assumptions. Ours are built on 15 years of designing structures that cannot fail. By embedding performance-based seismic design into AI-driven models, we transform fragility from generic to engineering-grade.

— Carlos Caramés Molero, Founder & Partner, Dynamis

That difference does not stay academic. It propagates straight into the indicators (re)insurers and lenders actually price on: PML for catastrophe risk capital on a critical-infrastructure corridor, AAL for technical premium across the operating life, EP curves for accumulation control across the Panama portfolio, SCR for Solvency II capital efficiency on the cedant side. Generic vulnerability functions cannot tell the difference between a metro line modelled as a single asset class and one whose damage curve is dominated by a specific viaduct interface. Ours can — because we engineered the interface.

Sources & references

1. Dynamis project portfolio: dynamisassociates.com
2. Metro de Panamá, Línea 3 — programme communications and procurement documentation, 2018–2024.
3. HPH Joint Venture (Hyundai Engineering & Construction · POSCO · Hanjin Heavy Industries), contractor consortium.
4. SENER Ingeniería y Sistemas, structural designer of record.
5. Jacobs Engineering Group, independent peer reviewer for the line's seismic and structural engineering.
6. Pedro Miguel and Limón fault systems, characterised in regional geological literature applicable to the Canal corridor.
7. Internal Dynamis design archive: 2022 hazard model and site response (peer-reviewed by Jacobs) and 2023 pounding analysis and damper design at the V8-6 interface — confidential, not public.