← Cases
South KoreaManufacturing2024–2025

Hallacast

AI quality prediction under the autonomous factory programme

Auto-parts caster Hallacast cut its defect rate from 10% to 6.5% with AI quality prediction under a government programme

WorkingCompany's own claim

Problem

Hallacast makes moulds and lightweight-material parts for vehicles, supplying Korean large companies and global carmakers. Its casting process carried defect and safety risks, and each worker could look after only one machine. Among Korean small manufacturers, upfront cost is the most cited barrier to AI: 44.2% named it in a government survey.

Approach

The company joined the SME ministry's autonomous factory programme in 2024 and applied it in two stages. In the first year it built automated lines 1 and 2 and piloted an AI quality-prediction system; in 2025 it added lines 3 and 4, a digital-twin simulation and stronger AI anomaly detection. Robots were placed in the casting process to reduce accident risk.

Results

10% → 6.5%defect rateCompany's own claim[1]
54 s → 47 scycle time on the electrical-parts lineCompany's own claim[1]
+15%productivity, with each worker now running two machines instead of oneCompany's own claim[1]
~90%accuracy of the AI quality prediction in 2025 (target 99%)Company's own claim[1]

Takeaway

Public funding spread the cost over two years and let the company start with a pilot before extending AI to more lines. The ministry's reported results combined the effects of automated lines and AI.

Categories

Business function
Production and quality
Innovation type
Process innovation
Evidence
Company's own claim

Sources

Source summaries

[1]전자신문23 Jul 2025

ET News reported the SME ministry's plan to expand its autonomous factory programme, with Hallacast as the lead example. Hallacast makes moulds and lightweight parts for future vehicles. The ministry reported the defect rate falling from 10% to 6.5%, cycle time from 54 to 47 seconds, productivity up 15% and AI prediction accuracy around 90%. ET News also cited a survey finding that only 1.7% of small manufacturers had adopted or planned AI.

[2]머니투데이24 Oct 2025

Money Today reported the ministry's AI-based smart manufacturing strategy, which aims to raise AI adoption among small manufacturers from about 1% to 10%. A survey found that 44.2% of small manufacturers saw upfront cost as the main barrier.

Full details are in the original sources.

Added 25 Oct 2026 · Updated 25 Oct 2026
Open case page ↗