ISA Batch Manufacturing Data Model Extension

Overview

An ISA-95/ISA-88 based manufacturing domain model for Cognite Data Fusion (CDF) that provides a complete, standards-compliant structure for batch and discrete manufacturing operations. Deployed as a Cognite Toolkit module, it defines spaces, containers, views, and a composed data model, along with SQL transformations, seed data, and a workflow to orchestrate data loading.

The model integrates two key international manufacturing standards: ISA-95 for the organizational hierarchy (Enterprise → Site → Area → ProcessCell → Unit → EquipmentModule) and ISA-88 for batch procedural control (Recipe → Procedure → UnitProcedure → Operation → Phase). It is optimized for industrial scale, high performance, and AI readiness — built around an asset hierarchy to simplify navigation, with AI-friendly naming across all views and properties. The model is intentionally a starting point: it contains placeholders and structures where projects can add properties, and add or remove views as required per customer.

Who is this for & when to use

Target personas: Delivery Engineer, Data Modeller, Process Engineer, CDF Administrator on a batch manufacturing or discrete manufacturing project.

Use this pack when:

• Setting up a new CDF project for manufacturing operations and you need an ISA-95/ISA-88 compliant data model without building it from scratch.
• You need to integrate organizational hierarchy, batch execution, production management, and process data into a unified, queryable structure.
• You want AI-ready views, CDM/IDM compliance, and cross-hierarchy navigation (ISA-95 organizational ↔ ISA-88 procedural) out of the box.
• Extending an existing CDF project with a standardised manufacturing, batch control, or work order data model.
• You want a validated reference to compare or evaluate your existing manufacturing data model using AI tooling.

When NOT to use this pack:

• If your project is oil & gas operations rather than manufacturing — use the CFIHOS Oil & Gas Data Model deployment pack instead.
• If your manufacturing domain does not use batch control and has no need for ISA-88 procedural hierarchy — this model will require significant simplification.

Why you need it

Without this pack, teams must design an ISA-95/ISA-88 compliant data model from scratch — mapping the full organizational hierarchy, procedural model, production management entities, and the cross-hierarchy relationships that connect them. Doing this correctly while implementing CDM/IDM extensions, configuring transformations, and wiring up dependency-aware workflows typically takes weeks.

This pack eliminates that effort by providing a production-validated, fully connected model that:

• Implements both ISA-95 and ISA-88 in a single unified model — organizational hierarchy, batch procedural control, production management, and process data integration are all pre-wired.
• Uses purpose-specific containers with targeted indexing for faster queries, ensuring smooth compatibility with CDF Search and Canvas.
• Delivers AI-friendly naming and descriptions on all views and properties, improving navigation accuracy and AI output quality.
• Includes dependency-aware workflow orchestration so transformation execution order is handled automatically, with retry logic and timeout protection built in.
• Ships with seed data and test CSV files so the model can be validated end-to-end without production source data.

What's included

Prerequisites

Verify all of the following before starting deployment:

• Cognite Toolkit version 0.8 or later is installed. Follow the setup instructions.
• A cdf.toml file exists in your project root. If missing, run cdf init and select Create toml file (required).
• Authentication is configured and verified — run cdf auth init and cdf auth verify using a local .env file. See the Toolkit authentication docs.
• Your CDF project has Data Modeling enabled.
• The CDF group used by Toolkit has the following capabilities: spaces:write, datamodels:write, transformations:write, workflows:write, datasets:write, rawdb:write.
• You have access to enable feature flags for your CDF project in Cognite Unleash (or can request it from your CDF administrator).

Deployment & installation

Choose your setup path

1. Existing Toolkit project

Use this path if you already have a local Toolkit project and only want to add the ISA Manufacturing data model.

a. Point cdf.toml at the Cognite Library

Long-time Toolkit users may still have [library.cognite] pointing at toolkit-data. Update it to the official library release:

[library.cognite]
url = "https://github.com/cognitedata/library/releases/download/latest/packages.zip"

b. Enable deployment packs in cdf.toml

In the same cdf.toml, ensure the alpha flag is set:

[alpha_flags]
deployment-pack = true

c. Add the ISA Manufacturing module

From your project directory:

cdf modules add -d isa_manufacturing_extension isa_manufacturing_extension_search

d. Build and deploy

Run in order:

cdf build
cdf deploy --dry-run
cdf deploy

After deployment, the ISA Manufacturing data model, containers, and views are available in your CDF environment.

Verify folder structure

Your project should include:

modules/
└── data_models/
    ├── isa_manufacturing_extension/
    └── isa_manufacturing_extension_search/

2. Starting from scratch

Use this path for a new Toolkit project with only the ISA Manufacturing data model (and the data models pack scaffolding Toolkit creates).

a. Initialize the project

In an empty directory:

cdf modules init .

b. Select ISA Manufacturing in the interactive UI

In the terminal:

• Choose Data models (data models that extend the core data model).
• Use Space to select ISA 88/95 Manufacturing Data Model template and ISA 88/95 Manufacturing Search Solution Model.
• Press Enter to confirm.

Toolkit downloads the library, creates the module under modules/data_models/isa_manufacturing_extension/, and updates config.dev.yaml / config.prod.yaml.

c. Build and deploy

cdf build
cdf deploy --dry-run
cdf deploy

After either path, the ISA Manufacturing data model, containers, and views are deployed to your CDF environment.

3. Enable the FDX_VIEW_SWITCHER feature flag

After deployment (both paths), enable the FDX_VIEW_SWITCHER feature flag in Cognite Unleash for your CDF project. This enables view navigation, relationship visualisation, and enhanced data model exploration in the CDF UI.

Verification

After deployment, perform the following checks:

1. Verify that the deployment completed successfully. The cdf deploy command should finish without errors and report successful deployment of spaces, containers, views, data models, transformations, and the workflow.

2. In Cognite Data Fusion, navigate to:
   Data Modeling → Spaces
   Verify that the sp_isa_manufacturing and sp_isa_instance spaces (or your configured names) both exist.

3. Navigate to:
   Data Modeling → Data Models
   Verify that the ISA Manufacturing data model is listed and can be opened.

4. Open the ISA Manufacturing data model and verify that key views are present, including: Enterprise, Site, Area, ProcessCell, Unit, EquipmentModule, Equipment, Recipe, Procedure, UnitProcedure, Operation, Phase, WorkOrder, Batch, ISATimeSeries, and ISAAsset.

5. Navigate to:
   Data Modeling → Containers
   Verify that the expected ISA Manufacturing containers have been deployed in the sp_isa_manufacturing space.

6. Navigate to:
   Workflows → wf_isa_manufacturing
   Confirm the workflow completed without errors. All transformation tasks should show a successful execution status. Note that the asset tree build task (isa_asset_tr) must complete before the ISA-95 overlay transformations run.

7. After the workflow completes, navigate to the Data Modeling Search UI and verify that entities such as ISAAsset, Equipment, WorkOrder, and ISATimeSeries contain instances and that cross-hierarchy relationships (e.g. Unit → Equipment → Phase → ISATimeSeries) are navigable.

The deployment is successful when both spaces exist, all views are accessible in CDF, the workflow has completed without errors, and instances are visible in the Search UI.

Configuration reference

Parameters are set in config.dev.yaml at the project root. Default values work for an initial test deployment.

Architecture

isa_manufacturing_extension/
├── default.config.yaml              # enterprise_space, instance_space, enterprise_dm_version, …
├── module.toml
├── data_modeling/
│   ├── containers/                  # 24 enterprise container definitions
│   │   ├── dm_dom_isa_manufacturing_Area.Container.yaml
│   │   ├── dm_dom_isa_manufacturing_Batch.Container.yaml
│   │   ├── ...
│   │   └── dm_dom_isa_manufacturing_WorkOrder.Container.yaml
│   ├── views/                       # 26 enterprise view definitions
│   │   ├── Area.view.yaml
│   │   ├── Batch.view.yaml
│   │   ├── ISAFile.view.yaml
│   │   ├── ISATimeSeries.view.yaml
│   │   ├── ...
│   │   └── WorkOrder.view.yaml
│   ├── dm_dom_isa_manufacturing.DataModel.yaml   # Enterprise (DOM) data model
│   ├── dm_dom_isa_manufacturing.Space.yaml       # Schema space ({{ enterprise_space }})
│   └── inst_isa_manufacturing.Space.yaml         # Instance space ({{ instance_space }})
├── data_sets/
│   └── isa_manufacturing.DataSet.yaml
├── locations/
│   └── loc_isa_manufacturing_edm.LocationFilter.yaml
├── raw/                             # RAW table definitions + optional seed CSVs
├── transformations/                 # tr_isa_* transformations (RAW → instance space)
│   ├── tr_isa_asset_all_to_area.Transformation.{sql,yaml}
│   ├── tr_isa_asset_all_to_enterprise.Transformation.{sql,yaml}
│   ├── ...
│   └── tr_isa_timeseries_all_to_isa_timeseries.Transformation.{sql,yaml}
├── workflows/
│   ├── wf_all_isa_raw_to_isa_manufacturing.Workflow.yaml
│   └── wf_all_isa_raw_to_isa_manufacturing.WorkflowVersion.yaml
└── files/                           # Sample PDFs for file-ingestion demos

The model integrates ISA-95 and ISA-88 into three interconnected hierarchies. The ISA-95 organizational hierarchy provides the physical structure (Enterprise through EquipmentModule). The ISA-88 procedural model defines batch execution (Recipe through Phase). ISA-95 Level 3 production management (ProductDefinition, ProductRequest, ProductSegment) bridges planning to execution. Cross-hierarchy relationships and ISATimeSeries connect all three.

ISA-95 Organizational Hierarchy (Levels 0–4)

ISA-88 Procedural Model

ISA-95 Level 3 Production Management

Supporting entities

Key design decisions

Dual-standard model. ISA-95 and ISA-88 are implemented in a single unified module rather than separate packs. This allows a WorkOrder to link directly to both a ProductRequest (ISA-95 Level 3) and a Batch / Phase (ISA-88), enabling end-to-end traceability without cross-module joins.

ISAAsset as the central navigation hub. Only ISAAsset implements CogniteAsset. All hierarchy nodes (Enterprise, Site, Area, ProcessCell, Unit, EquipmentModule, Equipment) are linked to ISAAsset rather than implementing CogniteAsset individually. This preserves a clean single-hierarchy navigation pattern in CDF applications and avoids UI conflicts in IndustryCanvas and Asset Explorer.

Intentional starting point. The module ships with placeholder structures and stub properties where projects are expected to add domain-specific content. Views can be extended with new properties, added or removed, and containers can be extended — without breaking the existing schema contract. This keeps the baseline deployable on day one while remaining adaptable to real customer data shapes.

Key cross-hierarchy navigation paths

• Production planning to batch execution: ProductDefinition → ProductRequest → WorkOrder → Batch → Phase → ProcessParameter
• Organizational to procedural: Site → Unit → Equipment → Phase → ProcessParameter → ISATimeSeries
• Production planning to process data: ProductSegment → ProcessParameter → Phase → ISATimeSeries
• Work management to execution: WorkOrder → Equipment → Phase → Batch → Recipe

Extending the module

To add a new entity: create a container in data_models/containers/ (naming convention: sp_isa_manufacturing_<Entity>.Container.yaml), create a matching view in data_models/views/<Entity>.view.yaml, add relevant implements: entries for cdf_cdm interfaces, and include the view in ISA_Manufacturing.DataModel.yaml. Add a corresponding *.Transformation.sql and *.Transformation.yaml in transformations/, then update workflows/wf_isa_manufacturing.Workflow.yaml with the new task and its dependencies.

Troubleshooting

For issues not covered here, contact your Cognite support team or post in #topic-deployment-packs on Slack.

Support & ownership