Madhava Venkatesh Raghavan is CTO and Co-Founder of TrusTrace, a leading platform for supply chain traceability within fashion and retail.

getty

Call it an example of the government stepping in to do the will of the people. As we speak, the European Commission is working to implement its Product Environmental Footprint (PEF) program, which, when it becomes law, will require companies to calculate and disclose the environmental impact of their products by tracing their origins through supply chains.

Such legislation is a victory for activists and consumer rights groups that have long pushed brands to operate more sustainably, especially fashion brands. According to widely accepted estimates, the fashion industry accounts for as much as 8% of global greenhouse gas emissions.

PEF will join a slew of regulations worldwide, including California’s Transparency in Supply Chains Act and Germany’s recently passed Supply Chain Due Diligence Act, forcing large companies to get a better handle on their supply chains. This will require technology solutions for supply chain traceability, plus a new way of thinking about sustainability.

Until now, companies have approached sustainability at the corporate level—a top-down push to act more responsibly and market products accordingly. But this is already dated thinking (and, frankly, something most companies already do). What’s now required—especially by regulators and consumers—is product-level sustainability.

Gathering sustainability data at the product level means knowing everything about every product and material a company handles. It requires much more granular, accurate data—at scale. Fashion brands command complex, intricate supply chains, including everything from dyes and yarns to fabrics and finished goods. For them, traceability is critical.

The Basics Of Traceability

The ability to precisely trace products and materials through a supply chain can help address many challenges:

• Traceability can lead to greater supply chain visibility, which can help anticipate and avoid the kind of disruptions we’ve seen recently.

• It supports proof of product provenance—knowing exactly where a product and its materials came from.

• It’s the basis for adhering to product claims and proving authenticity. A brand can’t say a sweater is made of 100% organic cotton unless it can trace the origins of that cotton.

• Traceability can result in social accountability, ensuring workers are treated fairly and improving lives in the process.

• And, in the case of the European Commission’s PEF program and similar laws, traceability is required to meet regulatory requirements.

There are certain supply chains in which traceability is well established. The highly regulated pharmaceutical industry, for example, is well experienced in traceability and the automated systems required to establish it.

The fashion industry is nothing like that. Fragmented, sprawling and with little-to-zero supply chain visibility, fashion companies face a daunting task not only in their need to meet new supply chain regulations but also in their efforts to streamline supply chains to support better business outcomes and customer experiences. Striving toward sustainability at the product level requires new thinking.

Product-level responsibility is a technology challenge that will require a full stack of highly scalable, accurate solutions, from artificial intelligence (AI) and machine learning to IoT tagging that joins the physical and digital worlds and blockchain-powered systems that create a trusted, evidence-based chain of custody. In upcoming articles, I’ll delve into each component of the traceability technology stack, but let’s start with AI—without which we can’t even get started.

AI As The Digital Bridge To Traceability

In my experience working with companies in the fashion industry, most of the data needed to move toward product-level sustainability and supply chain traceability is locked up in documents—paper and electronic. This can include everything from invoices that prove the chain of custody and social audit reports describing a facility’s workplace conditions to chemical tests for specific batches of materials. All this document data could be in different formats and languages.

On top of that, new digital-native data is starting to influence supply chains: satellite images, video data of facilities, employee interviews and testimonials and IoT data from machinery and energy meters. In short, traceability represents a complex data acquisition problem.

While we develop technologies that make existing trade, certification and testing processes natively digital, we need intermediate systems that can help digitize paper trails. This is where AI can come into play—not only because of its OCR-style ability to intelligently ingest data but also its ability to support a system that automatically performs data validation by correlating information from multiple sources to improve the overall data quality.

This digitization of “paper trails” to enable traceability includes the following steps:

Classification: First, the AI recognizes and identifies a document submitted somewhere in the supply chains and intelligently classifies it as, for example, a purchase order, facility audit or certification.

Object Extraction And Identification: Based on the document’s classification, AI identifies the key information based on metadata. For example, when processing invoices, the system captures information like buyer, seller, product, quantity and date of delivery. Similarly, digitizing a social audit might involve capturing parameters related to working conditions, fair wages, diversity and more.

Data Validation And Linking: Once the corresponding objects are extracted from the supporting documents, the data is validated and linked to other data within existing enterprise systems to allow companies to act on it, whether for forecasting, analytics, regulatory reporting or other requirements.

When supply chain traceability data is all digital, AI and machine learning can have ongoing roles. Supply chains are so complex and the available data so vast that AI will be needed for companies to act fast, such as when one partner’s sustainability can’t meet a brand’s standards and the supply chain must adapt and reconfigure through other partners to remain in compliance.

As the European Commission’s PEF program demonstrates, it soon won’t be enough for brands to simply prove sustainability; they’ll have to calculate in near real time how sustainable their products are by intelligently tracing their combined materials. This is the textbook definition of product-level responsibility.

Many fashion brands were already committed to sustainability and social responsibility before the world’s legislators became involved. That corporate commitment must now permeate brands at the product level. It’s a big challenge but AI is one way to meet it.


Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?