--- title: "Unify — Canonical Supply Chain Data Model for WMS, TMS & ERP | bluefabric" description: "bluefabric unifies fragmented WMS, TMS, ERP, EDI, spreadsheet, API, and legacy data into one canonical, self-describing supply chain data model — the master data foundation AI agents need to reason over operations across systems." url: https://bluefabric.ai/unify/ source: content/unify/index.html --- [bluefabric](/)/ [Product](/ingest/)/ Unify · Data Model Pillars [Ingest](/ingest/) [Enrich](/enrich/) [Unify](/unify/) [Context](/context/) [Calculate](/calculate/) [Use](/use/) Layer 3 of 5 · Unify # Turn fragmented data into one _self-describing operational model._ Your systems were not designed to speak the same language. **ERP sees the order. WMS sees the warehouse work. TMS sees the shipment. Spreadsheets see the exceptions. None of them sees the full chain.** bluefabric maps fragmented, flat, two-dimensional data into a common supply chain model built from 1,000+ real consulting projects — then exposes every object, relationship, and tool in a way AI agents can understand and use. Supply chain is not flat. Your AI data layer should not be either. [See bluefabric Live →](/demo/) 15-min walkthrough // objects · relationships · tools A row is not a supply chain ## Rows do not run supply chains. _Relationships do._ Most operational data arrives as tables. Orders Shipments Inventory Suppliers SKUs Costs Exceptions But supply chains do not operate as tables. **They operate as relationships.** An order connects to line items, inventory, warehouse activity, labor, shipments, carriers, suppliers, customer commitments, service risk, and cost impact. bluefabric turns disconnected records into a connected operational model that reflects how supply chains actually behave. Rows do not run supply chains. Relationships do. Orders & Inventory Logistics & Flow Suppliers & Customers Signals & Risk ![](https://pub-6d0b5b97762c4335b5b515672d21523f.r2.dev/img/technology/network-mesh-wire-structure-light-background.jpg) Built from 1,000+ supply chain projects ## Supply chain knowledge, _encoded into software._ The bluefabric data model is not a generic schema designed in a whiteboard session. It is built from 1,000+ supply chain consulting projects across warehouse operations, transport, inventory, fulfillment, supplier management, ERP, WMS, TMS, and execution improvement. That means the model already understands the objects, relationships, and edge cases that appear in real operations. What it is The object, named correctly. Order, SKU, shipment, inventory position, supplier, carrier, warehouse, exception, cost, demand signal, service risk — defined as supply chain entities, not generic rows. How it's named One vocabulary across systems. Different systems call the same thing by different names. bluefabric standardizes the language so agents never trip over \`cust\_no\` vs \`customer\_id\` vs \`BPNumber\`. How it relates Every connection, modelled. Orders connect to inventory. Inventory connects to fulfillment. Fulfillment connects to shipments. Shipments connect to carriers, costs, and customer commitments. ![Physical string-art network graph — nodes connected by stretched threads, the literal shape of relationships](https://pub-6d0b5b97762c4335b5b515672d21523f.r2.dev/img/technology/network-string-art-installation-grey.jpg) // every system contributes a piece Every system fills in a different part of the truth ## Every system contributes a piece. _bluefabric builds the chain._ No single system tells the full story. bluefabric brings the fragments together and fills the blanks in the right place — so the result is one connected model of how the operation works, not another copy of your data. ERP Order header, customer, **commercial terms**, invoice state. WMS Line items, pick status, pack activity, **stock location**, who handled the work. TMS Carrier, lane, tracking status, **ETA, dwell, delivery risk**. Files Exceptions, partner updates, **missing operational detail** nobody automated. Every system contributes a piece. bluefabric builds the chain. From flat data to graph relationships ## The value is not the record. _The value is the relationship._ bluefabric does more than standardize columns. It turns operational data into graph relationships so agents can reason across the chain instead of reading one record at a time. Supply chain graph traversal Customer → Order → Line Item → SKU → Inventory → Warehouse Activity → Shipment → Carrier → Delivery Commitment → Service Risk → Cost Impact Q 01 Which **customers** are affected if this shipment is late? Q 02 Which **supplier delay** is blocking priority orders? Q 03 Which **SKU shortage** is creating SLA exposure? Q 04 Which **warehouse constraint** is causing transport cost drift? These are not dashboard queries. They are supply chain questions. Fact shipment\_status = "DELAYED" Operational meaning **Customer SLA at risk.** Notify rep. Re-evaluate route. Fact inventory\_qty = 42 Operational meaning **Reserved · blocked · expired?** Can it actually fulfill demand? Fact order\_id = SO-4821 Operational meaning What other data does this order need to be **actually delivered**? Best practices encoded ## Systems store facts. _bluefabric adds operational meaning._ bluefabric does not only map data into objects. It brings supply chain knowledge into the model — **where each object belongs, how entities should relate, what attributes matter, which fields are usually missing, and what decisions depend on them.** A source system might store a shipment status. bluefabric understands whether that status creates customer risk. A WMS might store inventory quantity. bluefabric understands whether that inventory is available, blocked, reserved, damaged, expired, misplaced, or unable to fulfill demand. Systems store facts. bluefabric adds operational meaning. ![](https://pub-6d0b5b97762c4335b5b515672d21523f.r2.dev/img/technology/network-nodes-white-lines-dark-background.jpg) Designed for agents, not dashboards ## Dashboards need fields. _Agents need context._ Most data models were built for reporting. bluefabric is built for AI agents that need to reason, calculate, and act — so every object carries the context an agent needs to use it correctly. Complete Connected context across every system. Agents see the same model spanning ERP, WMS, TMS, spreadsheets, APIs, and legacy systems — no system left out. Deterministic Resolves against structure, not prompt interpretation. Questions land on structured entities, relationships, and verified data — the agent doesn't guess what a field means. Action-ready Facts connected to decisions. The model links facts to calculations, recommendations, and governed write-back — ready for an agent to act, not just answer. Where Unify sits ## Clean records become _connected context._ The common data model sits at the center of the bluefabric flow. It is where fragmented data becomes operational intelligence. 01 Ingest bring sources in → 02 Enrich clean · resolve · backfill → 03 · here Unify common data model → 04 Calculate trusted methods → 05 Use MCP · any agent Clean records become connected context. Connected context becomes useful AI. [Explore Trusted Calculations →](/calculate/) [Architecture deep-dive](/architecture/) ![](https://pub-6d0b5b97762c4335b5b515672d21523f.r2.dev/img/technology/global-network-earth-space.jpg) A complete operational model ## One model. One tool layer. _Any agent._ bluefabric turns ERP, WMS, TMS, spreadsheets, APIs, and legacy data into one common supply chain model built for agentic operations. **Not another flat export. Not another dashboard schema. Not another disconnected data warehouse table.** A better model creates better agents. [See bluefabric Live →](/demo/) [Next: Calculate →](/calculate/) [ ← Back: Layer 2 Enrich How bluefabric cleans, resolves, and backfills supply chain data before it lands in the model. ](/enrich/)[ Next: Layer 4 → Context · Company knowledge Turn SOPs, contracts, PDFs, terms, and tribal knowledge into agent-ready context layered on top of the model. ](/context/)