Manufacturing Traceability: Fighting Counterfeit Parts with AI | Roei Ganzarski

What if every manufactured part could be identified and traced using nothing more than a photograph?

In this episode of The Manufacturers Network Podcast, Lisa Ryan talks with Roei Ganzarski, CEO of Alitheon, about manufacturing traceability, counterfeit parts, quality control, AI, automation, and technology that gives physical objects something similar to their own unique fingerprint.

Roei's background includes Boeing, aviation, software, electric aircraft, and manufacturing technology. Throughout his career, he kept encountering the same problem: manufacturers rely on paperwork, serial numbers, barcodes, QR codes, labels, and tags to prove that a physical part is the part everyone thinks it is.

But what happens when the label falls off, a serial number is copied, a QR code is moved, or a counterfeit part looks almost identical to the original?

Can a Manufactured Part Have a Fingerprint?

Alitheon's technology, called FeaturePrint, uses photographs to identify individual physical items without adding a barcode, tag, sticker, RFID chip, or other marker.

Manufacturing processes produce parts within allowable tolerances, but seemingly identical parts still have tiny physical differences. FeaturePrint uses those variations to create what Roei describes as "biometrics for things."

That means a manufacturer can potentially identify whether a specific item has been seen before, trace it through production, or determine that an item isn't the one it claims to be.

Counterfeit Parts Are a Manufacturing and Safety Problem

Counterfeiting isn't limited to fake watches and designer bags.

Roei discusses counterfeit and gray-market products involving pharmaceuticals, automotive brake pads and airbags, healthcare products, military components, and other high-consequence applications.

For manufacturers, the risk extends in both directions.

Incoming counterfeit tooling, components, filters, or materials can affect production and quality. Outgoing counterfeit products can create safety risks, warranty problems, financial losses, and damage to brand reputation and customer trust.

Traceability Can Improve Manufacturing Quality

Traceability becomes especially valuable when something goes wrong.

Imagine parts begin drifting toward the edge of an acceptable tolerance. If individual parts can be traced through production, leaders may be able to identify what the affected parts have in common, such as a specific machine, production line, shift, or point in the process.

The same principle applies when products come back for warranty claims or service.

Instead of wondering whether there's a quality problem across the entire operation, better manufacturing traceability can narrow the investigation and give people better information for making decisions.

AI, Automation and Industry 4.0

As manufacturers pursue Industry 4.0, smart factories, AI, and greater automation, Roei argues that companies need visibility into more than what their machines are doing. They also need reliable information about the physical products moving through those machines.

Technology can provide data. Human beings still need to interpret it, question assumptions, investigate problems, and decide what to do.

For Roei, that makes critical thinking increasingly valuable as factories become more automated and data-driven.

How Should Manufacturers Evaluate AI Technology?

Manufacturers are being flooded with AI pitches. Roei recommends asking practical questions before investing:

• Does the technology solve an expensive, high-consequence problem?

• Can the vendor demonstrate it live using your products?

• Who owns the intellectual property behind the solution?

• Is the AI proprietary technology or primarily built on someone else's platform?

• Does it solve the physical problem or only manage the data around it?

• What happens if the label, barcode, identifier, or underlying assumption is wrong?

Roei also challenges manufacturers to distinguish between protecting data and protecting physical products. Blockchain may protect digital records, for example, but accurate data can still be connected to the wrong physical item if the identifier itself is compromised.

In This Episode

• Manufacturing traceability without barcodes, labels, or serial numbers

• How digital fingerprints can identify individual manufactured parts

• Counterfeit and gray-market risks in manufacturing

• Quality control and product serialization

• AI-powered physical item identification

• Supply chain traceability and authentication

• Warranty fraud and product returns

• Industry 4.0 and smart manufacturing

• Human judgment and critical thinking in automated factories

• How manufacturers should evaluate AI vendors and technology investments

• Why better data can help people make better manufacturing decisions

As manufacturing becomes more automated, knowing what a machine did is only part of the story. Manufacturers also need confidence that the physical part in front of them is exactly the part they think it is.

Guest: Roei Ganzarski

CEO, Alitheon

#Manufacturing #ManufacturingTechnology #AIinManufacturing #Traceability #CounterfeitParts #Industry40 #SmartManufacturing #QualityControl #SupplyChain #Automation #HumanJudgment #ManufacturersNetwork