Open Research Life Cycle Assessment

• Open Research Life Cycle Assessment (ORLCA) is an open-source repository that centralizes bespoke lifecycle inventory data tailored for research applications.
• It addresses data gaps by aggregating proprietary, incomplete, and specialized environmental inventories, thereby enhancing reproducibility and comparability.
• The platform promotes FAIR principles and supports digital twin integration, streamlining sustainable design decisions for complex research infrastructures.

Open Research Life Cycle Assessment (ORLCA) is a proposed Open Research Life Cycle Assessment repository: a first-of-its-kind open-source repository of LCA inventory specifically for research, but available for use by others beyond, designed to centralize collated general and bespoke LCA data for research contexts where conventional Life Cycle Assessment (LCA) databases are often insufficient (Wakeling et al., 10 Sep 2025). Its stated purpose is to make comprehensive LCAs more feasible, comparable, and reproducible for scientific experiments and research infrastructures by addressing recurring obstacles such as proprietary or unpublished data, missing inventories for bespoke components and specialized processes, and constraints of cost, time, training, and expertise. In this conception, ORLCA is not merely a store of background data; it is intended as open research infrastructure for environmentally responsible planning, operation, and decommissioning.

1. Concept and motivation

ORLCA is grounded in the standard LCA view that environmental impacts should be assessed across the full lifecycle of a product or system, from raw material extraction and manufacturing through use, disposal, and recycling. The underlying methodological reference points are ISO 14040 and ISO 14044, while the immediate problem is domain-specific: research infrastructures and experimental systems often depend on materials, components, and processes that are bespoke and therefore absent from existing inventories (Wakeling et al., 10 Sep 2025).

The proposal responds to a set of concrete deficiencies in current research LCA practice. Available data may be proprietary, unpublished, paywalled, outdated, geographically unsuitable, too coarse, or simply absent. Open and commercial databases remain useful for bulk materials and generic processes, but they often lack the level of detail needed for large facilities, accelerator subsystems, astronomy instrumentation, detector hardware, and related scientific assets. ORLCA is therefore framed as a repository that would provide unrestricted and sustainable access to data while reducing duplicated effort and lowering the barriers created by data accessibility, availability, financial costs, and expertise (Wakeling et al., 10 Sep 2025).

A central illustration is the ISIS-II Neutron and Muon Source case. In that study, the practitioner relied on proxies, engineering concepts, bills of materials, and later a commercial ecoinvent license, yet still could not obtain all data required for a full LCA. The paper uses this as evidence that even technically mature LCA workflows become difficult, expensive, slow, and incomplete when research-specific inventories are missing (Wakeling, 2024). A related example is the LCA of the GRAND experiment, which documented transparent assumptions and a mostly open data basis precisely because supplier data, process details, and end-of-life routes were incomplete (Vargas-Ibáñez et al., 2023).

This suggests that ORLCA is motivated less by dissatisfaction with LCA as a method than by a mismatch between existing data infrastructures and the epistemic structure of research systems. Research infrastructures are often novel, heterogeneous, and evolving; consequently, their environmental accounting depends on repositories capable of handling partial information, proxies, bespoke inventories, and iterative refinement.

2. Intended scope, users, and decision functions

The repository is intended to provide several classes of data products: research life cycle inventories, materials data, and LCA results, both simulated and measured (Wakeling et al., 10 Sep 2025). Its target use is broad. The initial strategic communities identified are particle physics, accelerator physics, and astronomy, fields characterized by large scientific facilities and existing sustainability-oriented strategic planning contexts such as the European Strategy for Particle Physics Update (ESPPU), SNOWMASS, and Science Vision for European Astronomy.

ORLCA is explicitly framed as support for decisions across the full lifecycle of research infrastructure and experiments. The relevant stages are planning and design, operation, and decommissioning / end-of-life. More specifically, the repository is expected to inform choices concerning material selection, energy consumption, waste management, and other sustainability practices. The claimed practical benefit is that researchers could assess material and operational impacts without building a dataset from scratch, potentially saving months or years of iterative sustainable design work and supporting environmental impact statements in funding applications (Wakeling et al., 10 Sep 2025).

This orientation aligns with published research-infrastructure LCAs that use environmental accounting early in design rather than retrospectively. The ISIS-II study describes an iterative and simplified LCA conducted during the feasibility and optioneering phase so that design options can still be influenced; its preliminary results identify buildings and shielding as likely hotspots and show how even first-order inventories can guide architectural choices (Wakeling, 2024). The GRAND study similarly used component-level inventory analysis to identify the antenna structure and battery as dominant contributors and translated those findings into recommendations on alloy choice, material reduction, battery lifetime, recycled materials, and transport (Vargas-Ibáñez et al., 2023).

The repository is also expected to support digital twin integration for pre-prototype environmental evaluation. A plausible implication is that ORLCA is intended to shift sustainability analysis earlier in the design cycle, before physical prototyping or irreversible procurement decisions.

3. FAIR alignment, platform design, and repository structure

ORLCA is described as being “designed to comply fully” with FAIR principles: findability, accessibility, interoperability, and reusability (Wakeling et al., 10 Sep 2025). The implementation path is conceptual rather than fully specified, but several concrete platform decisions are named.

For findability, the repository is proposed to be hosted on Zenodo, a general-purpose open repository run by CERN under OpenAIRE, with records organized through communities and each record assigned a persistent DOI. For accessibility, ORLCA is intended to be open-access and free, while still requiring appropriate licensing and attribution for contributed data. For interoperability, import and export are expected to use a common LCA format, with JSON-LD and ILCD given as examples, and compatibility with open-source or free software such as OpenLCA treated as a priority. For reusability, records are to be curated, versioned, documented, and peer reviewed; updated records receive a new unique DOI, preserving traceability and version control (Wakeling et al., 10 Sep 2025).

The repository’s operational content would begin with baseline data from existing open-source databases for bulk materials and generic processes, then expand via partner contributions, third-party data from relevant and intersectional fields, and relevant materials data. A notable requirement is completeness and stand-alone use: records should include all relevant supplementary data so that the repository is usable independently, especially for benchmarking between studies (Wakeling et al., 10 Sep 2025).

The following table summarizes the repository features stated for ORLCA.

FAIR dimension	Stated ORLCA mechanism	Named examples
Findability	Searchable records, community organization, persistent identifiers	Zenodo, communities, DOI
Accessibility	Open-access and free use, with licensing and attribution care	Zenodo hosting
Interoperability	Common LCA data exchange formats and software compatibility	JSON-LD, ILCD, OpenLCA
Reusability	Curation, versioning, documentation, peer review, DOI-based citation	quality standard document, new unique DOI

A related literature on interoperable LCA data models sharpens the meaning of interoperability in this setting. Work on ontology-based LCA argues that FAIR alone is insufficient because it does not define ontologies or software blueprints; semantic interoperability requires shared vocabularies, machine-readable metadata, and graph-based infrastructures capable of representing activities, flows, inventories, provenance, and impact categories in a way that multiple systems interpret consistently (Malek et al., 2024). This is directly relevant to ORLCA because its own interoperability ambitions depend not only on file formats but also on stable semantic structures.

4. Data ingestion, curation, governance, and quality control

The ORLCA proposal does not present a formal software schema, but it does specify an operational model for data collection and governance. Intended ingestion modes include data submission by users, ethical data scraping, data digitisation, experimental data collection, including via the Internet of Things (IoT), and responsible use of machine learning and artificial intelligence to assist collection and processing (Wakeling et al., 10 Sep 2025).

Governance is structured around a quality standard document defining accepted formats, detail levels, and uncertainty thresholds; community standards; a public Curation Policy; tutorials on how to collect and calculate LCA data; data submission guidelines; a peer-review process; citation/DOI assignment; and role-based permissions controlling who can review submissions, accept or decline them, edit metadata, and manage community settings. This architecture is meant to enforce quality control while preserving openness (Wakeling et al., 10 Sep 2025).

The repository’s treatment of incomplete information is also explicit. Where detailed inventory data are unavailable, ORLCA would supply high-level data with uncertainties as a default. The implied methodology is hybrid: begin with open databases for generic background data, supplement them with contributed primary and secondary data, use proxies only when necessary, and aim over time for more detailed and uncertainty-bounded inventories. This is consistent with the practical reality documented in research-infrastructure LCAs, where inventory analysis often begins from partial designs and proxy systems and becomes progressively refined as supplier data and engineering specifications mature (Wakeling, 2024).

Recent computational LCA research suggests a possible technical complement to these ingestion and curation mechanisms. One multimodal AI-agent workflow treats LCA as an information-retrieval and reasoning problem over public text and images, using a custom data abstraction, iterative interaction between an LCA Agent and a Stakeholders Agent, and public sources such as product pages, datasheets, certification sites, FCC reports, and iFixit teardown pages to construct LCI entries and estimate impacts (Zhang et al., 22 Jul 2025). That work is not an ORLCA framework formally, but it demonstrates how AI-assisted extraction from public sources could support ORLCA-style data collection and gap filling without eliminating the need for expert verification.

5. Relationship to open computational and semantic LCA ecosystems

Although ORLCA is introduced as a repository concept, its practical realization depends on a wider ecosystem of interoperable models, open tooling, and computational workflows. Several adjacent developments in the literature define this ecosystem.

First, ontology-centered LCA research argues that richer metadata, shared semantics, and graph databases are prerequisites for a reusable and machine-actionable LCA infrastructure. Examples cited in that literature include the consensus ontology model by Kuczenski et al., the LciO ontology by Meyer et al., the BONSAI ontology, the life cycle engineering ontology by Wilde et al., and graph-based systems such as LCIKG and PFKG. The same work emphasizes RDF, OWL, SPARQL, Neo4j, and Cypher, and presents a modular ingestion architecture in which workflow tables, metadata tables, agent tables, and reference tables are normalized and loaded into a graph database for semantic querying (Malek et al., 2024). In ORLCA terms, this supplies an interoperability backbone for representing provenance, activities, flows, and contextual metadata.

Second, open computational LCA for ICT argues that LCA artifacts should be treated as explicit, versioned model objects with dependency structure rather than opaque spreadsheet outputs. Its four requirements—explicit model lineage, clearly defined model scope, end-to-end traceability, and managed non-obsolescence—map closely onto ORLCA’s stated concerns with DOI-based persistence, versioning, traceability, and benchmarking (Corlay et al., 7 Apr 2026). The proposed mechanism of explicit dependency graphs and an open, versioned LCA-oriented repository indicates how ORLCA could evolve from a data repository into a broader model repository.

Third, open-source software packages make advanced open workflows operational. The lcpy package provides an open-source Python interface for static, Monte Carlo, dynamic, and parametric LCA/LCC, with integration points to brightway 2.0, Pymoo, SALib, and prospective LCA tooling. Its hierarchical structure of Main Process, Sub-processes, and Sub-sub-processes shows one way ORLCA-hosted data could be embedded in uncertainty-aware and optimization-oriented research pipelines (Gkousis et al., 16 Jun 2025).

Fourth, domain-specific large-scale data generation frameworks show what ORLCA-aligned resources may look like when a field reaches higher maturity. The CRYSTAL framework for organic chemicals is described as open-source code and versioned datasets, transparent, modular, and scalable, and it generates more than 110,000 transparent LCI datasets for more than 70,000 organic chemicals from molecular structure using retrosynthesis, rule-based and machine-learned process modeling, and pathway-resolved inventory generation (Chen et al., 15 Mar 2026). While CRYSTAL addresses chemicals rather than research infrastructure, it demonstrates an ORLCA-like transition from manual, opaque, case-by-case inventory building to inspectable and collaboratively improvable computational infrastructure.

Taken together, these developments indicate that ORLCA belongs to a broader shift in LCA practice: from isolated studies toward open repositories, semantically structured metadata, versioned computational models, and reusable pipelines.

6. Limitations, unresolved issues, and future directions

ORLCA remains a concept, not yet a fully implemented repository (Wakeling et al., 10 Sep 2025). The proposal identifies several unresolved problems. Many required datasets do not yet exist; some data will remain approximate or proxy-based; and it will be difficult to guarantee adequate granularity, accuracy, and regional representativeness. Data come from many locations and formats, so a multi-faceted approach will be required for compilation and standardization. The review burden is also open: the proposal recommends that review be kept as simple as possible, but it does not define the full operational workflow in detail.

A major open question is self-sustainability and longevity. The proposal suggests that after initial infrastructure is built, the repository could continue with minimal upkeep through community contribution, but explicitly states that this is not guaranteed. The project is also seeking funding for personnel and implementation, so institutional support and community buy-in are preconditions for realization (Wakeling et al., 10 Sep 2025).

Licensing and legal issues remain underdetermined. ORLCA must ensure attribution of contributed data, appropriate licensing, and legal compliance, but the open-source license for ORLCA itself is deferred to a reconnaissance stage. This is significant because repository openness does not remove obligations around contributed datasets, proprietary interfaces, or re-distribution constraints.

The literature also clarifies several misconceptions. One misconception is that FAIR compliance by itself solves interoperability; ontology-based work argues that FAIR is a governance principle, whereas semantic interoperability requires formal vocabularies and explicit metadata structures (Malek et al., 2024). Another misconception is that openness eliminates modeling uncertainty; open computational LCA for ICT argues that versioning, scope control, lineage, and traceability are necessary precisely because models remain provisional, database-sensitive, and vulnerable to silent obsolescence (Corlay et al., 7 Apr 2026). A third misconception is that automation can replace expert judgment. AI-assisted sustainability assessment demonstrates substantial speed and accuracy gains, but its authors explicitly caution against interpreting such tools as fully replacing expert verification in high-stakes accounting (Zhang et al., 22 Jul 2025).

The next steps named for ORLCA are concrete: determine the best open-source license; perform reconnaissance on data collection methods; establish detailed quality standards; develop the community contribution framework; populate the repository with baseline and contributed data; and expand beyond the initial target disciplines if successful (Wakeling et al., 10 Sep 2025). This suggests a staged trajectory in which ORLCA first consolidates disjointed efforts in research LCA, then becomes a platform for cross-checking, benchmarking, and sustainable design support across a wider set of scientific domains.