A weekly briefing for the industrial economy

The Operations
Brief

By Bill Murphy, powered by Colony Spark

 

September 25, 2026

A new look for The Operations Brief

We refreshed the design to match our new website, so the newsletter and the site now read as one publication. What you get from it hasn’t changed: three sections every week, written for the people running manufacturing and distribution operations. Thanks for reading.

Most operators can trace product reasonably well inside their own four walls. The trouble starts at the handoff, when a supplier’s records have to make sense inside someone else’s ERP, WMS, and quality system.

This issue is about that handoff: why it’s becoming part of the supplier relationship, what it looks like at a real receiving dock, and a practical way to find where your own evidence chain breaks.

This week’s question

Could your receiving team trace one supplier lot from last month, PO to production use, without digging through email?

Hit reply with yes, no, or “depends on the supplier.” One word is enough. I’ll share what readers say in a future issue.

Bill

01The Shift

The Traceability Handshake Is Becoming a Supplier Requirement

Traceability evidence now has to survive the boundary between companies.

A shipment can reach the dock before its usable traceability record does. The product is physically there. The lot references, certificates, event records, and receiving data that would let the next operator accept it are not, or they arrive in a form the receiving system can’t use. So the team rebuilds the trail by hand from inboxes, portals, and disconnected systems before it can decide whether to accept, hold, investigate, or escalate.

Most companies can trace at least part of a product inside their own environment. The harder problem starts when that history has to cross into another company’s ERP, WMS, QMS, MES, or receiving process. Item identifiers differ. Lot and serial references don’t match. Certificates travel separately from the shipment. Event records use different structures.

NIST gave this problem a current technical foundation on September 9. NIST IR 8536 lays out a manufacturing traceability meta-framework for building a continuous, time-ordered provenance chain across separate supply chain ecosystems. It relies on common data patterns, interoperable interfaces, cryptographically verifiable links, and selective disclosure, so companies can share the evidence traceability requires without exposing proprietary operating data or depending on a central repository.

The framework matters less than what it signals for operators. The shift is from visibility to usability. A supplier can hold the right information and still fail the handoff if the receiver can’t connect it to the product, lot, shipment, event, or decision inside its own workflow. Traceability is becoming part of the supplier handshake, judged alongside the product, the ASN, and the delivery date.

A test for your receiving team

Pick one supplier lot from last month. Can your team connect it to its PO, shipment, certificate, receipt, inspection, and downstream use without rebuilding the story from email threads and separate systems? If the answer takes a day of digging, the handshake is already broken.

Holding the evidence isn’t the same as handing it over in a form the next company can use.

For Cencora, this isn’t a future architecture problem. A truck can reach the dock before the data required to receive its product.

02From the Floor

The Truck Is at the Dock. The Data Isn’t.

Gregg Gorniak of Cencora on receiving product that arrives before its evidence.

Gregg Gorniak is Vice President of Manufacturer Operations and Data Services and Secure Supply Chain Lead at Cencora, which receives pharmaceutical shipments from manufacturer partners. Under the Drug Supply Chain Security Act, Cencora needs the matching serialized data before it can receive covered product and take ownership of it. In a Q&A with Pharmaceutical Commerce, Gorniak described how his team compares purchase orders, ASNs, and expected delivery windows against incoming EPCIS data.

“I don’t want to be reactive when the truck arrives at my dock and then I can’t break the seal and open the truck and get the product off.”

Gregg Gorniak

VP, Manufacturer Operations and Data Services, Cencora. Pharmaceutical Commerce, Sept. 2026

His fix is an early warning. A red, amber, and green view flags suppliers whose EPCIS data hasn’t arrived 12 to 24 hours before the truck is due. That window gives Cencora and the manufacturer time to repair the data handoff, whether the cause was a system glitch or a file nobody pushed, before the shipment reaches receiving. Without it, the truck is physically present and the product is operationally unavailable.

Gorniak also draws a line between compliance and a usable handoff. A portal, PDF, or spreadsheet can satisfy a minimum access requirement and still leave the receiver stuck, because a PDF can’t be ingested into the system that validates the product. In his framing, a portal makes a supplier compliant. It doesn’t make it a good partner.

The takeaway

DSCSA is specific to pharmaceutical distribution, not a rule for every plant or warehouse. The operating lesson travels anyway: the quality and timing of the traceability record is now part of the supplier relationship, not an afterthought to the product and the ASN.

03The Stack

The Supplier Traceability Reconciliation Agent

One place to rebuild the supplier-to-production evidence chain, with humans reviewing only the exceptions.

Supplier evidence arrives through ASNs, certificates, packing documents, emails, portals, and shared event records. Internal evidence sits across the PO, the ERP receipt, the WMS scan, the QMS inspection, inventory movements, and BOM or production consumption. Today the buyer or quality lead spends most of the effort matching identifiers and putting events in order before investigating the actual discrepancy.

This agent does the assembly work and hands a person only the missing or conflicting links. It’s a practical system pattern, not a named product you can buy off the shelf.

What it reads

 

Supplier side: ASN, packing list, certificates, and supporting documents.

 

Purchasing and ERP: PO line, item, revision, expected quantity, supplier master, receipts, and inventory transactions.

 

Warehouse and quality: WMS scans and location moves, plus QMS inspection, disposition, and nonconformance records.

 

Production: MES or BOM consumption records where available.

 

Optional: EPCIS events, when both supplier and receiver use the standard.

How it runs, in five steps

1. Select. The operator picks a component, lot, serial number, shipment, or receipt.

2. Normalize. The agent extracts and normalizes the supplier and internal identifiers.

3. Sequence. It orders the available events from supplier shipment through receipt, inspection, inventory, and production use.

4. Compare. It checks supplier claims against internal records and flags missing, duplicate, late, or conflicting evidence.

5. Decide. The buyer or quality lead reviews the exceptions and approves the next action.

What it returns

 

A time-ordered traceability packet with links back to every source record.

 

A broken-link list that shows the first missing or conflicting handoff.

 

The affected material, inventory, WIP, or assemblies the current data supports.

 

An evidence view for the buyer, receiving lead, or quality lead.

 

A drafted supplier evidence request, held for human approval.

What it looks like on a real receipt

A certificate references supplier lot A417. The ASN uses a shipment-level identifier, and the ERP receipt records an internal lot. The agent links all three records, checks quantity and revision, and finds that the inspection result is tied to the wrong internal lot. It sends that exception to quality. It doesn’t release or hold the material on its own.

Two paths to build it

Path 1: Internal build

Combine document extraction, an event store, read connectors to ERP, WMS, QMS, and MES, identifier-mapping rules, optional EPCIS semantics, and a human exception queue. Begin with one supplier, one component family, and read-only access.

Path 2: ERP or implementation partner

Build the same flow around the supplier-document, lot, receipt, quality, and trace-reporting modules already in place. The operating team keeps ownership of identifier mappings, exception rules, and release authority.

Guardrails

 

It reconciles the evidence that exists. It can’t make false source data true or prove that an undocumented event happened.

 

It doesn’t release material, change a system of record, issue a hold, or contact a supplier without human approval.

 

Every conclusion keeps its source record and a confidence or gap status.

 

EPCIS gives partners a shared event language. It doesn’t replace ERP, WMS, QMS, MES, or supplier source records.

 

Cost depends on connector access, document variability, event volume, and partner participation, so there’s no universal price. Start read-only and narrow.

The first useful version doesn’t need to trace the whole enterprise. It needs to show one supplier and one receiving team exactly where their shared evidence chain breaks.

Before you go

Could your receiving team trace one supplier lot from last month, PO to production use, without digging through email?

Reply with yes, no, or “depends on the supplier.” If it’s a no, tell me where the trail breaks. I read every reply. Bill

The Operations Brief

By Bill Murphy, powered by Colony Spark

A weekly briefing on the decisions, people, and systems shaping manufacturing and distribution.

 

Colony Spark, 170 Mason Street, Greenwich, Connecticut 06830, United States

© 2026 Colony Spark. All rights reserved.