Validation at the ingest boundary
Each incoming batch is checked against its declared schema, units and code lists before it is accepted. Failures are quarantined with a reason, and the original is retained unchanged beside the standardised record.
GCRI · Foundry
Foundry 的数据工程
Data engineering decides whether evidence from different sectors can be combined at all. In this field it covers ingestion, validation, standardisation, provenance and stewardship: source passports, schemas and ontologies, frozen data cuts, and lineage that lets a correction reach every product built on the corrected source.
02申请
What data engineering, interoperability and stewardship does here.
Data engineering, interoperability and stewardship cover the pipelines, standards and governance that make data usable and trustworthy. They determine whether information from different sectors can be combined for joint 风险评估. In data and compute and methods and modelling, sound data engineering allows hazard, exposure and financial data to be linked consistently. It supports transparent and repeatable analysis of interconnected risk. The canonical description is on STEM Commons.
Frozen data cuts for reproducible products
Indices and loss estimates are computed from versioned, immutable snapshots, so a product can be re-run months later from exactly the data that produced it.
Correction impact through lineage
When a source owner issues a correction, the lineage graph lists every derived table, map and index that depends on it, and those products are marked stale until recomputed.
03方法与证据
How Data engineering at Foundry is modelled, measured and tested.
模型、其假设、数据和不确定性，以科学家或工程师会核查的形式呈现。结构遵循已发表的方法。所取数值仅为示意，不是研究发现。
模型、方程与假设
The model structure used for Data engineering at Foundry, the assumptions it makes and how each is tested.
要素 i 的风险
Ri = f(Hi, Ei, Vi, Ci)危害、风险暴露、脆弱性和应对能力预期年度损失
EAL = ∑k pk · Lk对年概率为 pk耦合系统
xt+1 = A xt + B ut + εtA carries the couplings, u the shock, ε the error| 假设 | 为何制作 | 如何检验 |
|---|---|---|
| 在一次事件中，危害与暴露相互独立 | 使损失积分保持可计算 | 与基于联合记录拟合的 copula 进行比较 |
| 耦合在工作点附近是线性的 | 让冲击通过 A 传播 | 在线性范围之外施加扰动并比较 |
| 脆弱性曲线可以在不同地点之间迁移使用 | 很少有本地曲线可用 | 在有观测损失的地方据此进行验证 |
- 输入
- 带溯源信息的版本化数据集
- 成果
- 分布，绝不是单一数字
- 代码
- 连同模型卡和测试一起发布
- 复现
- 相同的数据、代码、环境和随机种子
FAIR 指导原则 · GO FAIR 和 FORCE11W3C PROV · 万维网联盟关于一致处理不确定性的指南 · IPCCCEOS 分析就绪数据 · 地球观测卫星委员会
04数据
该方法使用的输入。
Datasets and records used by data engineering, interoperability and stewardship on this platform, with the role each plays and the label its values carry. Sources are listed in 来源登记册.
| 数据 | 角色 | 标签 |
|---|---|---|
| Source passports | Owner, licence, cadence, authority class and limits | 已声明 |
| Schemas and ontology shapes | Validation at ingest | 已声明 |
| Pipeline run logs | Lineage and failure analysis | 观测 |
| Published statistics and registers | Authoritative reference values | 已发布统计 |
| Catalogue metadata in DCAT and STAC | Discovery and reuse | 已声明 |
05流程
流程如何运行。
Process explorer · Process
Data engineering at Foundry: Build, sign and accept
10 steps·4 个阶段·2 道关口
流程
Data engineering at Foundry: Build, sign and accept
获准的任务成为周期项目；构建版本经过固定、验证和签名，全球风险论坛（Global Risks Forum，GRF）记录的是用于演练的验收，而不是部署许可。
01 请求与来源
- 01已准入的任务成为周期项目记录 · 宇宙 到 Foundry已准入的任务成为附决策案例的周期项目。
- 02对数据来源与资源进行资格审查步骤 · Global Centre for Risk and Innovation (GCRI)GCRI 对数据来源与资源提供进行资格审查，并构建情景模型。提供不等于分配。
02 构建与验证
- 03来源快照记录 · NXS-FAC （蓝图、构建与签署）快照固定了提交、依赖项与生成器版本。
- 04构建合同记录 · NXS-FAC（蓝图、构建与签署）构建合同声明制品类型、环境与可复现性范围。
- 05整合与构建模型 · GCRIGCRI 整合软件与数据、签署构建版本，并测试安全性与互操作性。
- 06验证套件关口 · NXS-FAC（蓝图、构建与签署）必需检查与建议性检查相互分开。可复现性要求固定输入、声明非确定性并进行独立重建。
03 签署与准入
- 07签署确切的制品步骤 · NXS-FAC（蓝图、构建与签署）NXS-FAC 通过限定范围的签名者身份进行构建、证明并请求签名。它不持有任何密钥材料。
- 08资料包准入步骤 · NXS-NXQ （软件、模型与能力生命周期）NXS-NXQ 接收已签名的软件包，并保存将任务、数据类别、评估套件与有效期绑定在一起的准入证据。
04 接受
- 09核心构建报告记录 · GCRI核心构建报告是通过或不通过的记录。Universe 组装核心，并持有演练的决定权。
- 10G3 · 构建被接受用于演练关口 · Global Risks ForumGRF 记录构建版本是否被接受用于演练。已签名的构建版本或候选发布版本并不意味着获准部署到生产环境。
计划年度 · NXS-FAC · UniverseDesigned programme process, not a record of operation. No cycle has run and every programme is at service level R0. A gate is a programme decision, never a public-authority approval, certification or deployment authorisation. The record order follows the Builds and Releases room lists of NXS-FAC.
06验证
结果如何核查。
Pipelines are tested with automated quality checks, schema validation and lineage audits. Reconciliation against source 系统 confirms completeness.
| 核查 | 如何开展 |
|---|---|
| Reconciliation with the source | Row counts, totals and keys are compared with the source system after each load |
| Schema and unit validation | Records are tested against shapes, controlled vocabularies and CF or SI units before acceptance |
| Lineage audit | A sample of published figures is traced back to the original records and the transformation steps are re-run |
| Licence check | Every source and derived product carries a licence that permits its intended use |
Incompatible standards, missing metadata and restrictive licences block integration. Legacy systems and undocumented transformations introduce silent errors.
08管理责任
由谁维护该方法。
数据与算力 stewards data engineering, interoperability and stewardship for the Consortium and reviews how it is applied across platforms.
09已关联
本页在整个联合体中通向何处。
10问题
问题与解答。
What are the limits of data engineering, interoperability and stewardship in this field?
Incompatible standards, missing metadata and restrictive licences block integration. Legacy systems and undocumented transformations introduce silent errors.
结果如何核查？
Pipelines are tested with automated quality checks, schema validation and lineage audits. Reconciliation against source systems confirms completeness.
本页面上的数值是真实的吗？
The data names, roles and checks are real descriptions of practice. The canvas shows a design of the method’s use, not a run; no result is reported here.
该方法还应用于何处？
维持社会运转的 规范方法页面 列出应用该方法的每个平台。
报告贡献
从本页贡献证据。
报告 is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
GCRIFoundryFoundry 构建与准备度
- 电子邮件网关 GW-GCRI从本页发送证据、更正、案例、方法、审查、翻译或概况投稿。它会带上本页标签送达，由负责的编辑接收。
- Zenodo开放存缴存缴并提交到社区用您的 ORCID iD 在 Zenodo 上发布您自己的记录，然后提交到那里的 Reports 社区。由策展人审核；DOI 仍归您所有。
- 报告GCRI查看所有贡献方式贡献类型、各自的评审方式、您获得的署名、许可协议，以及绝不接受的内容。
切勿发送个人数据、敏感基础设施细节或您无权分享的任何内容。贡献在使用前会经过审查；发送贡献不赋予署名权或编辑权，赞助也绝不能换取研究结论。
法律
法律文件。
本网站上的地图、数据和图表仅作为背景和一般信息展示。它们不是建议、预测或预警；信息的局限见免责声明和数据说明。