← All tracks

Track 01 · Security

Decision Advantage at Sea

Turning fragmented maritime signals into decision advantage, strategically and operationally.

SaabSigma Vertex

Every challenge follows the same structure. Each track has a shared theme. Within it, each partner company has an arena: a real problem from their operations, a concrete build assignment, and a defined deliverable. Pick one arena and build for it. Whatever you build in one arena will be instructive to every partner in that track.

Background

Nordic naval operations depend on a system of systems: submarines, surface vessels, coastal radars, underwater sensors, and naval bases. The data that describes their configuration, status, and health is scattered across legacy PDFs, OEM systems, PLM/ILS tools, and spreadsheets. The result is the same failure at two altitudes: leadership cannot see fleet-level risk, and engineers cannot trust document-level detail. Both problems are decision problems, and both are solvable with AI on top of the data that already exists.

Choose your arena

Arena A · Saab

Fleet Resilience Engine

The problem

Strategic questions like "which assets are single points of failure for coastal surveillance?" or "what happens if this base closes for a week?" cannot be answered today without weeks of manual analysis across disconnected systems. By the time the answer arrives, the situation has changed.

Build

A fleet resilience engine. Ingest assets (platforms, coastal sensors, bases, key nodes) with attributes and dependency links; compute per-asset risk scores from health, logistics, and network impact; visualise risk on a map or graph; run what-if scenarios (base closed for 7 days, radar chain segment down, spare part shortage) with impact metrics and natural-language explanations; and suggest a ranked list of resilience actions for the next 30–90 days.

Deliverable

A working web or tablet prototype where a user can see fleet-level risk, run a what-if scenario, and read a ranked, explained action list.

Data sources

Synthetic, non-classified JSON/CSV dataset from Saab — a small notional Nordic naval force with assets, dependency links, health and maintenance events, spares, missions, and scenario presets.

Arena B · Saab

Engineering Co-pilot

The problem

Engineers lose hours locating the correct document revision for a specific hull and manually producing inspection reports. Refits slow down, and the risk grows that outdated information is used onboard.

Build

An engineering co-pilot. Let a user pick a hull, compartment, system, and task; surface the correct procedures, drawings, limits, and safety notes; answer natural-language questions ("What tests are required after replacing this module?"); detect configuration issues such as revision mismatches; and draft work and inspection reports from task context and user notes.

Deliverable

A working web or tablet prototype where a user can navigate to a specific task, get the correct verified documentation, and generate a draft inspection report.

Data sources

Synthetic, non-classified JSON/CSV dataset from Saab — a notional vessel with compartments, systems, technical documents, tasks, configuration rules, historical issues, and report templates.

Shared requirement for all teams

Your architecture must address shipyard and dockyard reality: intermittent connectivity and security constraints. Include an architecture sketch in your pitch showing how the system behaves when the network is not there.

Pitch

A short pitch covering the end user, where in the operation your solution deploys, and what it would take to run in production.

Data sources (all arenas)

Both arenas: optional starter notebooks (Python/TypeScript) with a minimal API, a context pack on naval environments and constraints, and access to Saab mentors on-site for questions on realism and deployment architecture.

Resources

Datasets, APIs and starter links for this track are added closer to the date.