OntoCrafter is a semantic pipeline for digital twins of ceramic materials. Structured lab data (ELN/LIMS) and process fragments mined from the literature flow through a shared, four-layer ontology (OCO) into a queryable, validatable knowledge representation. In short: an ELN records what you did; the ontology connects it to what is known — and makes it answerable.
Contents
- About OntoCrafter & OCO
- What makes OCO different — the ten design principles
- The seven-tier mechanistic-explanation skeleton
- Compliance, value chain & Digital Product Passports
- The methodology paper
- The DILEMA-K project
- Platform MaterialDigital
- Numberland & CeramicsExplorer
- Access, licensing & contact
About OntoCrafter & OCO
What is OntoCrafter?
OntoCrafter is an ontology-engineering platform that integrates materials-science knowledge across three industrial perspectives — materials research, EU regulatory compliance, and supply-chain / Digital Product Passport (DPP, Manufacturing-X). Instead of building a separate ontology for every material class or regulation, it uses one shared four-layer architecture.
What is a “semantic pipeline”?
It is the end-to-end process that turns heterogeneous materials data into a structured, machine-interpretable knowledge representation you can query and validate — the poster’s “semantic pipeline for digital twins of ceramic materials.” It has three stages:
- Ingestion — structured lab data (ELN/LIMS, spreadsheets, measurements) and literature/patent SIPOC fragments.
- Semantic model — the OCO ontology (the four-layer architecture from the methodology paper) gives every input a shared, formal meaning (OWL classes, properties, SHACL rules) instead of leaving it as tables or free text.
- Tooling — OntoCrafter populates and validates the model, producing the digital twin.
It is semantic because the data carries formal meaning that machines can reason over, and a pipeline because inputs flow through defined stages into a result you can traverse via SPARQL and verify via SHACL.
Is there a live example of the semantic pipeline?
Yes — CeramicsExplorer (ceramicsexplorer.de) is a public application. It ingests hundreds of thousands of heterogeneous sources — patents, papers, funding awards, policy news, events — and turns them into structured, cross-referenced material knowledge: profiles for 90+ ceramic substances with linked patents and synonyms, research-institute and market tracking, and a funding radar. It is a live demonstration of the ingestion + semantic-classification side of the pipeline at scale — heterogeneous documents in, queryable structured knowledge out. (It is a discovery and monitoring publication rather than the deep per-material digital twin, but it runs nearly on the same semantic backbone.)
What is “digisuite”?
digisuite is Numberland’s branded name for the four-component toolkit that builds material digital twins — which is exactly the semantic pipeline above. Its four building blocks are the methodology paper (the architecture), the OCO ontology (the semantic model), the SIPOC-Extractor (literature ingestion), and the software stack (OntoCrafter, which populates and validates the model). In short, “digisuite” and “the semantic pipeline for ceramic digital twins” refer to the same four building blocks.
What is the OntoCrafter Ceramics Ontology (OCO)?
OCO is the reference implementation of the OntoCrafter architecture, focused on functional ceramics. Version 0.94 comprises 5,196 classes across 44 modules, 167,348 OWL axioms, 1,172 SHACL rules, and 5,920 Neumann symmetry constraints. It is published under a persistent identifier at w3id.org/oco and documented at ontocrafter.eu.
What is the four-layer architecture (L0–L3)?
OCO separates concerns into four independently versioned, modularly exchangeable layers:
- L0 — Bridges to standards: mappings to dozens of external vocabularies across nine domains (foundation ontologies, units & time, crystallography, LCA, industry data spaces, EU regulation, governance, international frameworks, identifiers).
- L1 — Material-agnostic skeleton: specimen, equipment, measurement, identifier, provenance — shared across all material classes.
- L2 — Material & compliance detail: swapped out when the material class or regulation changes, leaving the layers below untouched.
- L3 — Categorical reasoning: competency questions encoded as SHACL, plus categorical axioms ready for an opt-in OWL 2 DL reasoner.
Can OCO be extended to other (ceramic) materials?
Yes — and easily, which is a direct consequence of the layering. L0 (bridges), L1 (the material-agnostic lab/ELN level) and L3 (reasoning) are shared by every material; only L2 carries material-specific knowledge. L2 deliberately combines two kinds of modules: stable, reusable carriers — crystal symmetry with the 230 space groups, the tensor / Neumann engine, and the seven-tier mechanistic skeleton (universal physics for crystalline materials) — and material-specific catalogues — concrete materials, defect chemistry, phase regions — that are extended or swapped per material.
- For another ceramic (a new functional ceramic beyond BNT-BT and NiCuZn ferrite), the seven-tier ceramics skeleton applies unchanged; you mainly add the new material’s classes and per-tier instance content (phases, defects, coordination polyhedra).
- For a different material domain (metals, polymers, pharmaceutics), you keep L0/L1/L3 and contribute only a new L2 module.
Either way the rest of the ontology — and its reasoning — stays intact. This is the Portability principle, and it is what dissolves the ~70 % laboratory-workflow duplication that plagued earlier consortium projects, where every project re-modelled the same lab concepts.
What are the three industrial views on OCO?
One shared ontology serves three consumers: Materials Research (mechanistic depth — seven explanatory layers from symmetry to bonding), Manufacturing & DPP (value chain and Digital Product Passport via AAS/IEC 63278, Catena-X, Manufacturing-X), and EU Compliance (regulatory modules such as CSRD/ESRS, LCA, CSDDD, CBAM, AI Act, SSbD, PPWR). Competency questions across all three views are tagged to one of ten reasoning areas, so the reasoning scope is auditable per area.
What tools does OntoCrafter provide?
Two main tools: the OCO-Workbench / OntoCrafter software (a GUI for project management, graph visualization, class definition, plain-language SHACL authoring, and ABox population) and the SIPOC-Extractor, which mines process fragments (Supplier–Input–Process–Output–Customer) from publications and patents.
How does data get into OCO? (the two ingestion paths)
Two paths, used separately or together: (A) the Material Software Stack — structured lab data from ELN, LIMS, spreadsheets, or measurement exports; and (B) the SIPOC-Extractor — process fragments mined from the scientific literature. Both populate the OCO ABox (instances). In short: an ELN records what you did; the ontology connects it to what is known — and makes it answerable.
How exactly does structured lab data (ELN / LIMS) get into OCO?
OCO’s Level 1 (L1) is a material-agnostic LIMS/ELN level, deliberately designed to be ELN-/LIMS-importable. Lab records — samples, equipment, measurements, process steps, provenance — are written as ABox instances against L1’s material-agnostic vocabulary (plus the L0 bridges), so the same import works whether the sample is a ceramic, a metal or a polymer. Rather than writing a separate importer per ELN vendor, OCO bridges once to the ELN-Filetype standard — a cross-vendor, RO-Crate-based export format adopted by 14 ELN/LIMS systems — which covers the whole set without per-vendor mappings. A thin application profile (oco_eln_profile.ttl) bundles exactly the L0/L1 modules an ELN/LIMS consumer needs. An LIMS, for example, typically loads oco-investigation, oco-process, oco-equipment and oco-measurement and skips the rest.
What makes L1 “lab-ready”? (four operational features)
Four cross-cutting features make L1 work for real laboratory-data systems:
- Data Confidence Levels (DCL 0–5): a mandatory annotation on every measurement or extracted value, from Unknown (0) to Validated (5) — raw entries, peer-reviewed values and gold-standard data coexist in one graph and can be ranked.
- ExperimentStatus lifecycle: planned → in-progress → completed → archived — for any LIMS that distinguishes intent from execution.
- GUM uncertainty fields: combined standard uncertainty per JCGM 100:2008, attached to the existing measurement-uncertainty classes rather than a parallel hierarchy.
- LiteratureExtraction as a first-class
prov:Activity: extraction from publications/patents is a tracked activity with its own verification status, kept distinct from lab data.
L1 also carries the ISA lab-notebook skeleton (Investigation · Study · Assay · Sample · Batch · SampleState · Provenance). Because L1 is a separate, material-agnostic artefact, a sister project (metals, polymers, pharmaceutics) imports it unchanged and only replaces L2 — which dissolves the ~70 % workflow duplication seen across earlier consortium projects.
How do own experiments and literature fit together in one graph?
Through the Equality principle: laboratory experiments and published-literature extractions are semantically equivalent ABox contributors. Both are modelled as prov:Activity with the same provenance, quality and uncertainty annotations — distinguished only by their activity class (Investigation vs. LiteratureExtraction) and verification status, not by a separate vocabulary. So a query like “what is the d₃₃ of BNT-BT?” returns both validated in-house measurements and extracted literature values, ranked by data-confidence level, without the schema preferring one source a priori. On the poster this is the left ingestion path (Material Software Stack — ELN/LIMS, spreadsheets, measurement data); the SIPOC-Extractor is the right path — and both populate the same OCO ABox.
How exactly does the SIPOC-Extractor path (literature) get into OCO?
The SIPOC-Extractor is a pipeline that mines process fragments from scientific literature and patents and writes them into the OCO ABox against the same L1 vocabulary the lab path uses. The insight behind it: no single paper describes a complete process — so instead of demanding complete experiment records, data are ingested as SIPOC fragments (Supplier · Input · Process · Output · Customer). Fragments from many papers are then assembled into a complete laboratory recipe. Each extraction is recorded as a first-class LiteratureExtraction activity with its own provenance and verification status, so it never gets silently merged with lab data. On the poster this is Input B. On Numberland’s platforms this literature-extraction capability is branded .synthesize. — “distilling from thousands of documents the knowledge you lack today” — and on CeramicsExplorer the SIPOC extraction is shown under the .synthesize. section.
Why the SIPOC-fragment model? (handling incomplete data)
Real-world experiment documentation is never complete — and in literature and patents the incompleteness is often deliberate (competitive or patent strategy). A schema that mandated complete fields would either reject most real-world data or force fabricated placeholder values — both unacceptable. The SIPOC granularity, combined with OCO’s three-tier identifier hierarchy (mandatory / recommended / optional), accommodates incomplete records without breaking the schema. This is what makes OCO practically usable for its two dominant data sources — own lab data and literature extraction.
What does “Supplier–Input–Process–Output–Customer” mean here?
SIPOC is a quality-management framing applied to each experimental fragment: who supplied the inputs (Supplier), what went in (Input), what was done (Process), what came out (Output), and who consumes it (Customer). A paper — or a patent — is treated as a SIPOC fragment of one or more experiments, using exactly the same provenance pattern as an in-house experiment (the Equality principle). That is why property queries return lab measurements and literature-extracted values side by side, ranked by confidence.
Is the SIPOC-Extractor itself part of the published ontology?
No. The SIPOC-Extractor and the OCO-Workbench are companion software that use the OCO TBox as their schema. The ontology defines the target data model — SIPOC fragments, LiteratureExtraction activities, data-confidence levels — and the extractor populates it.
What is the “digital twin” OntoCrafter produces?
A fully structured, queryable and validatable knowledge representation of a ceramic material system — from raw process parameters to regulatory compliance — traversable via SPARQL and verifiable via SHACL.
What makes OCO different — the ten design principles
What makes OCO different from other materials ontologies?
OCO’s architecture is driven by ten engineering design principles (drawn from platform strategy, quality management/SIPOC, and materials science) rather than by a requirements checklist. Crucially, five of them go beyond the standard materials-ontology requirements canon and are OCO’s main differentiators: Adaptability, Equality, Compatibility, Authoritativeness, and Facetedness. In short: OCO is built like an engineering platform — modular, workflow-anchored, purpose-bound and reasoner-ready — with a mechanistic depth that most vocabularies lack.
What are the ten design principles?
- Modularity — classes and properties are organized into reusable modules combined per consumer (like the construction catalogues of automotive platforms); a consumer who needs only equipment and method vocabulary imports two modules, not the whole ontology.
- Adaptability — external ontologies are isolated in the L0 bridge level and cached locally with their versions; when an upstream ontology releases a new version only the bridge file changes, nothing in OCO breaks, and builds stay reproducible. Still-consolidating standards (EU Construction Products Regulation, Critical Raw Materials Act, DPP-XBRL) are tracked via provisional bridges.
- Interoperability — an ontology is a vocabulary embedded in a workflow, so OCO anchors concretely in the open-source materials stack: ELN/LIMS, machine-learning datasets (Croissant), materials databases (OPTIMADE), and provenance tooling (PROV-O).
- Purpose — scope is bound to a published catalogue of competency questions, each tagged with an area (tensor symmetry, phase state, route, lifecycle, …) and, where possible, backed by an executable SPARQL test. This prevents “dead classes”.
- Equality — every record is a SIPOC fragment (Supplier–Input–Process–Output–Customer), and a scientific publication is treated identically as a SIPOC fragment of one or more experiments. The same provenance pattern applies to both, removing the artificial line between raw experimental data and published results (this is what enables the SIPOC-Extractor).
- Compatibility — OCO is the architectural successor of two earlier PMD Phase-1 ontologies (KnowNow for LTCC multilayer ceramics, SmaDi for smart/piezo materials) and the EU-regulatory peer of the Manufacturing-X ecosystem (AAS asset administration shells, Catena-X, sector DPPs). Both backward and forward compatibility are achieved through bridges rather than re-modelling.
- Functionality — OCO is organized into four functional layers (L0 bridges, L1 lab-notebook/ELN-LIMS-ready, L2 material-specific, L3 reasoning), each addressing a different consumer concern. Maintenance stays local: a PMDco update touches only L0; a new material adds only L2.
- Authoritativeness — for functional ceramics OCO models the full mechanistic-explanation chain (local structure & bonding → symmetry → energy landscape → thermodynamic phase regions → kinetics → microstructure → defect chemistry → tensor-component symmetry) via a seven-tier skeleton, so it can answer why a property holds — e.g. “which piezoelectric coefficients are non-zero for BNT-BT at room temperature near the morphotropic phase boundary?” The pattern (Neumann’s principle) generalizes to all crystalline materials — metals and organic crystals included.
- Facetedness — every material parameter is classified along three independent axes simultaneously: role (state, response, transport, structure, statistical, topological, fit), reference (fundamental, material, interface, defect, measurement, specimen, process, environment), and layer (atomic, crystalline, microstructural, mesoscopic, macroscopic). The axes are reasoner-classified, so SPARQL can navigate any subset of facets without bespoke joins.
- Portability — the material-specific level (L2) can be replaced without touching L0, L1, or L3. A metallurgy ontology adopting the same L0/L1/L3 need only contribute its own L2 module to obtain a fully reasoner-ready ontology for its domain.
Which of the principles set OCO apart from the standard requirements canon?
Five principles are not captured by the standard materials-ontology requirements canon and reflect real consumer use-cases rather than a survey methodology: Adaptability (bridge-isolation plus external version cache for upstream drift), Equality (experiments and publications unified as prov:Activity — the prerequisite for the SIPOC-Extractor and cross-paper aggregation), Compatibility (ingesting predecessor ontologies verbatim instead of re-modelling), Authoritativeness (the full seven-tier mechanistic-explanation chain), and Facetedness (concrete multi-axis, reasoner-navigable classification). The other five (modularity, interoperability, purpose, functionality, portability) align with the canon but are realized with engineering rather than compliance intent.
The seven-tier mechanistic-explanation skeleton
What is the seven-tier mechanistic-explanation skeleton?
It is the internal organisation of OCO’s material-specific level (L2): a fixed stack of seven explanation tiers that links a material’s composition all the way to its measured property, so the causal chain is queryable from the same knowledge graph as the property value itself. This is what distinguishes an authoritative materials ontology — one that can explain why a property holds — from a vocabulary that only describes. The skeleton is generic: it applies to every crystalline ionic oxide (ferroelectrics, magnetics, ion conductors, high-Tc superconductors). BNT-BT was only the pilot material; NiCuZn ferrite, BaTiO₃, yttria-stabilised zirconia and related domains reuse the same seven tiers unchanged — only the per-tier instance content differs.
What are the seven tiers?
Each tier answers one fundamental materials-science question:
- Symmetry — which crystal structure and symmetry operations? Point-group theory (Neumann’s principle) decides which tensor components are non-zero; subgroup relations drive phase transitions; Wyckoff positions give dopant and defect sites.
- Energy / DFT — which electronic structure, phonon modes, Born effective charges? First-principles basis for intrinsic properties without empirical fitting; soft-mode analysis identifies phase-transition mechanisms.
- Thermo / CALPHAD — which phases are stable at (T, p, x)? Sublattice models and Redlich–Kister polynomials give Gibbs-energy surfaces → sintering windows, miscibility gaps, morphotropic phase boundaries.
- Kinetics — how fast do diffusion, reaction, sintering, switching and aging proceed? Fick, Arrhenius, JMAK, Hillert grain growth, Cahn–Hilliard, domain-wall mobility.
- Microstructure — which grain size, texture and grain-boundary character? Polycrystalline averaging via Voigt–Reuss–Hill, Mori–Tanaka, Hashin–Shtrikman bounds; Hall–Petch coupling to mechanical response.
- Defect chemistry — which point defects, in which concentrations, with which charge compensation? Kröger–Vink notation, Brouwer diagrams, donor/acceptor/iso-valent/amphoteric doping strategies.
- Bonding chemistry — which electron configuration, hybridisation and bonding character (ionic/covalent/metallic)? Pauling rules, HSAB classification, crystal-field theory, lone-pair stereo-activity.
How is the explanation chain kept consistent?
The chain is auditable: cross-tier references are formalised as seven SHACL shapes that flag incomplete annotations (as warnings, not hard errors) — e.g. a morphotropic phase boundary (Tier 3 Thermo) must reference a domain-state set (Tier 1 Symmetry); a Brouwer diagram (Tier 6 Defect chemistry) must reference a chemical-potential diagram (Tier 3 Thermo). The BNT-BT pilot demonstrates it works end-to-end: the query “why does BNT-BT achieve d₃₃ ≈ 580 pC/N at the morphotropic phase boundary?” traverses all seven tiers along these annotations.
Compliance, value chain & Digital Product Passports
How does OCO handle EU regulation and the value chain?
Through a second, independent classification axis — the audience axis — that runs orthogonally to the L0–L3 levels. Every module is marked material, compliance, or both. The material-science modules are organised by level; the compliance / value-chain modules cut across levels (each carries classes on several of L1, L2 and L3 at once). That is the architectural point: the same modular primitive absorbs the EU-driven convergence of materials, manufacturing, supply-chain and sustainability data without restructuring the material core — compliance is architecture, not appendix. A consumer who needs only the material core can ignore the compliance audience entirely; one building a Manufacturing-X pipeline loads both.
Which compliance / value-chain modules exist?
Fifteen modules carry the compliance audience — eleven compliance-pure plus four dual-audience (shared with the material side):
- Compliance-pure (11): LCA (Life Cycle Assessment), CSRD/ESRS reporting, CSDDD (supply-chain due diligence), PPWR (packaging & packaging waste), CBAM (carbon-border adjustment), R2R (right-to-repair), the EU AI Act, Manufacturing-X, regulated substances, recycling, and SSbD (Safe-and-Sustainable-by-Design).
- Dual-audience (4):
format(industrial data formats),odrl(W3C ODRL policy + Verifiable Credentials trust layer),time-event(W3C Time), andautomation(SiLA 2.0 laboratory automation).
Each has its own L1/L2/L3 internal structure and is anchored to the relevant standard — e.g. LCA to ISO 14040/44/67, EN 15804+A2 and EU PEF (bridged to ecoinvent, BONSAI); CSRD to the EFRAG XBRL taxonomy; CSDDD to UN Guiding Principles and ILO.
What is the Digital Product Passport (DPP) pattern in OCO?
The DPP is the joining concept that ties the compliance modules together. OCO models two complementary kinds:
- Material-DPPs (from
oco-lca) — per-material passport profiles attached to a specific material class (e.g. a BNT-BT passport). - Sector-DPPs (from
oco-mfgx) — per-industry profiles tied to specific EU regulations: Battery DPP, Construction-Products DPP (under CPR 2024/3110), and a Generic-ESPR DPP.
A concrete product inherits from both via multiple inheritance: a BNT-BT actuator inside a battery assembly is simultaneously a BNT-BT-DPP and a Battery-DPP, so it satisfies the material passport and the sector passport at the same time.
How does OCO connect to Manufacturing-X and Catena-X?
Via the oco-mfgx module, which anchors OCO in the Manufacturing-X identifier and traceability infrastructure: AAS asset administration shells (IEC 63278), Catena-X BPN identifiers and aspect models (e.g. CX-0010 Quality, CX-0146 Demand & Capacity), IPCC climate factors, and the sector-DPPs above. Because this is done through bridges (the Adaptability principle), OCO remains the EU-regulatory peer of the Manufacturing-X ecosystem while its material core stays untouched.
The methodology paper
Is there a scientific paper behind OntoCrafter?
Yes. Pannek, T. & Grond, W. (2026): “A Multi-Level Architecture for Reusable Materials Ontologies — The OntoCrafter Ceramics Ontology (OCO) as Reference Implementation.” arXiv:2606.14814, CC-BY 4.0. It introduces the layered architecture (abstraction levels L0–L3 and a material-vs-compliance audience axis) and a seven-tier mechanistic-explanation skeleton (symmetry, energy/DFT, thermo/CALPHAD, …), with OCO as the reference implementation.
The DILEMA-K project
What is DILEMA-K?
DILEMA-K — Digital Lifecycle of Lead-Free Piezoelectric and Ferrite High-Performance Ceramics — is a research project funded by Germany’s Federal Ministry BMFTR (March 2025 – February 2029) within Platform MaterialDigital 3 (grant no. 13XP5228C). OntoCrafter / OCO is the semantic data pipeline and ontology built within this project.
Who is in the DILEMA-K consortium?
CeramTec GmbH (lead), plus Aixacct Systems, Ernst-Abbe-Hochschule Jena, Fraunhofer IKTS, HappyOps, KMS Technology Center, Numberland, SONOTEC, TRIDELTA Weichferrite, TU Berlin, and VIA Electronic. Numberland contributes the semantic data pipeline and ontology.
Which materials does DILEMA-K focus on?
Lead-free piezoelectric and ferrite high-performance ceramics — the reference material systems are BNT-BT (bismuth-sodium-titanate / barium-titanate) and NiCuZn ferrite.
Platform MaterialDigital
What is Platform MaterialDigital (PMD)?
PMD is a German research initiative for the FAIR (Findable, Accessible, Interoperable, Reusable) digital representation of materials across their lifecycle. It uses ontologies and standardized workflows for interoperability while preserving each partner’s data ownership. It is funded by the BMFTR and carried by six founding institutions (BAM, Fraunhofer IWM, KIT, IWT, Max Planck Institute, FIZ Karlsruhe).
What does “MaterialDigital 3” mean?
PMD is funded in sequential phases (MaterialDigital 1, 2, 3, and PMD-X). DILEMA-K is one of the projects in the third funding phase (MaterialDigital 3).
Numberland & CeramicsExplorer
Who develops OntoCrafter?
Numberland — a German engineering firm founded in 1996 in Kulmbach, specialised in machine reading and text analytics of technical documents (its “TextAnalyticsFactory”). Dr. Wolfgang Grond is the owner; Thomas Pannek is CIO. Numberland provides the semantic data pipeline and ontology for DILEMA-K.
What is CeramicsExplorer?
CeramicsExplorer (ceramicsexplorer.de) is a Numberland publication — “Navigating ceramics, glasses and functional materials.” It aggregates and indexes materials data: 360,000+ documents (patents, papers, funding awards, policy news, events), profiles for 90+ ceramic substances, research-institute tracking, market barometers, and a funding radar. Editor-in-chief: Dr. Wolfgang Grond.
What do Numberland’s dotted keywords (.navigate., .ideate., .synthesize., …) mean?
They are the modules of Numberland’s TextAnalyticsFactory, its machine text-analytics suite; each is styled with surrounding dots:
- .navigate. — market strategy in the face of societal and technological change
- .ideate. — query expert-authored documents when the right expert isn’t available
- .remember. — surface internal knowledge (“if we knew what we know”)
- .explore. — technological market analyses, including complex terminology
- .compare. — benchmark external-vs-external or external-vs-internal
- .locate. — assign documents to European regions
- .identify. — account for data as an intangible asset
- .change. — align a business model with megatrends
- .synthesize. — distil the knowledge you lack from thousands of documents (this is the SIPOC-extraction / knowledge-building capability behind OCO)
- .combine. / .blend. — combine several modules, or build a fully custom solution
On CeramicsExplorer these appear as navigation sections (.navigate., .explore., .locate., .compare., .synthesize., .change., …) applied specifically to ceramics, glasses and functional materials.
Access, licensing & contact
What license applies to OCO?
A three-tier model: the L0 bridges are CC-BY; the public L1 skeleton is CC-BY-SA; the L2/L3 and compliance modules are proprietary.
Where can I access OCO?
OCO is documented at ontocrafter.eu and resolvable via its persistent identifier w3id.org/oco; the methodology paper is on arXiv (2606.14814). The public portion (the L0 bridges and the L1 skeleton, excluding the supplier module) is openly available.
How can I work with OntoCrafter? (three collaboration paths)
There are three concrete ways to collaborate, matched to the layer you care about:
- L0 + L1 — research project / consortium. If your organisation already has a materials-science ontology in another domain (polymers, metals, batteries, …), you adopt OCO’s shared foundation — the L0 bridges plus the material-agnostic L1 level — and connect your own L2 domain module; together we validate that the architecture holds. (CTA on the site: “Architektur-Dialog beginnen.”)
- L2 — industrial customer / new domain. If you need coverage for a material class OCO does not yet model and lack in-house ontology-engineering capacity, Numberland develops a modular L2 for you that plugs into the OCO framework — an engineering service. (CTA: “Projekt anfragen.”)
- L3 — reviewer / academic validation. If you are an ontology engineer, journal reviewer or researcher, you are invited to give critical feedback on the architecture and validation methods — with no obligation. (CTA: “Feedback geben.”)
How do I get in touch, and what should the email subject be?
Write to ontocrafter@numberland.com (Dr. W. Grond / T. Pannek; German or English; typically answered within two business days). To route your request, tag the subject line:
- OCO-Bridge — integrating an existing ontology or data format
- OCO-Modul — L2 development for a material class
- Werkzeug — early access to the tools
- Architektur — academic discussion, pre-prints, review
- Allgemein — anything else