SpainPublic sector2018 – ongoingGeneral Treasury of Social Security
Predictive models for contribution fraud and payment risk
Spain's social-security treasury used predictive models alongside monitoring of 1.5 million companies
Problem
Contribution fraud reduced social-security income. The treasury needed to identify suspicious employers and anticipate payment problems across a large administrative dataset.
Approach
TGSS equipped data scientists with AI tools to define behavioral patterns and predictive models. Risk alerts supported investigations into fictitious employment. Other models assessed payment-default risk and potential insolvency.
Results
Takeaway
The monitoring figures described coverage, rather than detection accuracy or savings attributable to AI.
Categories
- Size
- Large enterprise
- Industry
- Public sector
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Spain's social-security administration described operational predictive models and risk alerts at TGSS. It reported monitoring data on 1.5 million companies and 18.6 million active workers.
Full details are in the original sources.
