GFEMS Wins Innovation Award for Forced Labor Risk Detection Tool
GFEMS is excited to share that we have won in the “Prosperity” category at the 2020 Society for International Development- Washington Chapter Innovation Competition. Our entry, “Automated Decision Support Tool for Forced Labor Risk Detection” was judged by two separate panels of career nonprofit and international development professionals.
In developing the tool, GFEMS set out to solve the challenge companies, investors, authorities, or other stakeholders face in identifying the location(s) of forced labor in large, complex supply chains. With no existing viable tools developed that are both sustainable and effective, identifying forced labor in global supply chains has been nearly impossible. While numerous supply-chain risk assessment tools exist, the Fund undertook this project because existing tools suffer from one or more of the following shortcomings:
- They rely on qualitative information that is self-reported, expensive to collect and/or difficult to compare — all of which limits accuracy and scalability;
- Assessments are mainly restricted to tier-1 suppliers, so the vast majority of suppliers are excluded;
- They mostly provide high-level assessments of risk at a country level, which is insufficiently precise to enable meaningful action.
The GFEMS team, led by Senior Data Scientists Shannon Stewart, developed a novel decision support tool that predicts the risk of forced labor at the company level with about 84% accuracy. It uses data that is collected passively by governments and operates without participation of any of the firms on whom data is collected.
The tool is intended as a first-pass screening tool for use by corporate social responsibility and procurement professionals, investors, regulatory enforcers, and other stakeholders like NGO watchdog groups. It is not intended to replace these functions, rather it prioritizes due diligence efforts and stretches the impact of every dollar invested in cleaning up supply chains. It can elevate responsible manufacturing businesses in both access to markets and capital, and, in turn, support sustainable livelihoods for their workforces. Ultimately, the goal is to benefit the estimated 16 million people who are victims of forced labor within private-sector supply chains and to prevent more vulnerable people from becoming victims.
The Fund is currently gathering feedback on the tool from industry experts, and intends to release a refined version as an open source project. GFEMS has successfully demonstrated that there is reliable, detectable signal of forced labor risk in operational data. This proof of concept has encouraged at least one supply chain risk platform to begin development on a tool that operates on public data, combined with data that may be available only to them.
Moving forward, the Fund will work directly with companies who wish to implement a similar process. By opening our thinking to the public, GFEMS aims to inspire companies to take a new look at their data and how it fits with the broader context of industrial operations and to develop analogous tools that work for them.