Open-data capability demonstration · Zero-Emission Aviation
Mapping the UK Zero-Emission Flight Ecosystem
The UK zero-emission-flight sector has strong policy signals and funding, but the data on who is building what, who funds whom, and which technologies gate which pathway sits scattered across press releases, programme pages and reports. This is a worked, fully open demonstration of the data engineering that turns that fragmentation into a single, queryable, validated view, built on real named entities across the hydrogen and electric aviation ecosystem.
Entities
45
real, sourced actors, standards and technologies
Triples
482
with 8 provenanced quantities and 14 citable sources
Validation
0
SHACL violations; answers 6 competency questions
The Challenge
The UK is decarbonising aviation through the Jet Zero Strategy and significant public and private investment, from ATI FlyZero and the DfT-funded, CPC-led Zero Emission Flight Infrastructure (ZEFI) programme to demonstrators by ZeroAvia, Cranfield Aerospace Solutions, GKN Aerospace and Rolls-Royce. Yet the ecosystem remains fragmented across data sources, stakeholders and infrastructure, which limits the ability to see cross-sector relationships, track innovation and funding, and identify the end-to-end hydrogen pathways to zero-emission flight.
That fragmentation is exactly the problem a coordination tool has to solve. Here we show the underlying data engineering done in the open: a typed entity model, a controlled vocabulary, and machine-readable validation so that only stated relationships are represented and none are inferred.
The ecosystem, as one validated graph
Every node below is a real, sourced entity, and every edge a stated relationship: a funder funding a programme, a company developing a technology, a partner on a project, an aircraft demonstrating at an airport. The result is a single view that a policy maker, investor or airport planner can interrogate, the ecosystem-mapping primitive behind any “single view” coordination platform.
What the validation enforces
A graph is only a coordination tool if it can be trusted. The SHACL shapes make three guarantees, the same discipline a real coordination platform needs so its links are provided and validated, never guessed.
| # | Rule | What it guarantees |
|---|---|---|
| 1 | Every entity is labelled and typed | All 45 entities carry a human-readable label and one of nine ecosystem types (organisation, airport, programme, project, funder, body, alliance, standard, technology). |
| 2 | Maturity is controlled and sourced | Every technology carries either a maturity band from a controlled SKOS vocabulary, or a reified Technology Readiness Level assessment (integer 1 to 9) that is dated and derived from a named source. Maturity is never an unsourced literal. |
| 3 | Every quantity carries a unit and a source | Funding figures, capacities and projections are reified as quantities with a numeric value, a unit and provenance to a citable source, following the PECO emissions-provenance pattern (PROV-O plus QUDT). |
| 4 | Referential integrity on every relationship | The object of every relationship must be a declared entity of the correct type. Only stated links are represented; none are inferred. A negative test injects a dangling edge and confirms the validator catches it. |
What the graph contains
| Type | Count | Examples |
|---|---|---|
| Organisations | 15 | ZeroAvia, Cranfield Aerospace, GKN, Rolls-Royce, Airbus, easyJet, Loganair and more |
| Projects | 6 | Project NAPKIN, HyFlyer II, H2GEAR, H2FlyGHT, Project Fresson, Project Acorn |
| Programmes | 3 | ZEFI, ATI FlyZero, the Jet Zero Strategy |
| Funders | 3 | Department for Transport, Aerospace Technology Institute, UKRI Future Flight Challenge |
| Airports | 4 | Heathrow, London City, Bristol, Glasgow |
| Standards | 3 | ISO 19880-1 (gaseous H2 fuelling), ISO 13985 (LH2 tanks), CAA aerodrome safety |
| Technologies | 7 | Electrolysis, liquefaction, LH2 storage, fuel-cell powertrain, hydrogen combustion, refuelling |
The technology entities are linked by a feeds-into chain that models the physical hydrogen pathway, each stage carrying an indicative, dated maturity. That is the pathways-and-dependencies primitive, in miniature:
Stage 1
Production
electrolysis
Stage 2
Liquefaction
& distribution
Stage 3
Storage
cryogenic, on-airport
Stage 4
Refuelling
aircraft
Stage 5
Propulsion
fuel cell / turbine
Provenance-first: every number carries its source
A coordination tool is only as trustworthy as its evidence, so maturity and every quantity are modelled as first-class, sourced claims rather than bare literals, following the PECO emissions-provenance pattern (PROV-O plus QUDT). One competency question the graph answers, “which provenanced quantities does the graph hold, and from which source?”, returns the funding landscape below, each figure traceable to a primary source.
| Programme or project | Figure | What it is | Source |
|---|---|---|---|
| January 2021 green-aviation package | £84.6m | Total public + industry (ATI Programme) | gov.uk, 2021 |
| GKN H2GEAR | £54.4m | Total (£27.2m ATI grant + industry) | gov.uk, 2021 |
| ZeroAvia HyFlyer II | £24.6m | Total (£12.3m ATI grant + industry) | gov.uk, 2021 |
| GKN H2FlyGHT | £44m | 2 MW cryogenic hydrogen-electric propulsion | GKN, 2024 |
| Hydrogen in Aviation alliance | £34bn/yr | Projected UK economic benefit by 2050 | HIA / Rolls-Royce, 2023 |
The technology layer is modelled the same way: a reified Technology Readiness Level assessment carries an integer, an assessment date and a citation (for example, ATI FlyZero’s TRL 3 for the cryogenic liquid-hydrogen fuel system), so the graph can answer “which technologies gate a pathway because they are below TRL 6?” from evidence rather than assertion.
First of kind, built on the literature
A structured review conducted for this work found no published ontology or knowledge graph dedicated to hydrogen or zero-emission aviation, and no formal ontology model of Technology Readiness Level at all. Rather than invent from scratch, the design composes established work: the two-axis actors-and-value-chain structure of the HOLY hydrogen-market ontology (ISWC 2023); the provenance discipline of PECO; the continuant/occurrent separation of the Open Energy Ontology; the funding vocabulary of DINGO; and the standard W3C building blocks PROV-O, SKOS, QUDT and Dublin Core. It is engineered to the Linked Open Terms methodology, scoped and tested by competency questions, and published FAIR and open. Full citations are in the repository README.
Why it matters
The three primitives shown here, a typed stakeholder graph, a controlled-vocabulary technology model with sourced maturity, and referential-integrity validation, are the data spine of any “single view” coordination tool for a fragmented sector. Building them in the open, on real named entities, is how we demonstrate the capability rather than assert it. This is an independent open reference dataset, not a deliverable for any organisation named within it, and it is fully reproducible.
Open source and reproducible
The dataset, ontology, SHACL shapes, build-and-validate pipeline and an interactive explorer (built to WCAG 2.2 AA, with Mermaid diagrams and a keyboard-accessible table alternative to the network view) are published open source (CC BY 4.0) via Open Ontologies. Indicative maturity values are compiled from public roadmaps and statements, each dated, and are not an authoritative assessment.
View the code and data on GitHub