POSCO E&C debuts AI that automates ready-mix production and predicts strength before 28 days

POSCO E&C and SHLab built an AI system that predicts slump and strength, tunes mixes to KS limits, and logs every tweak. It steadies quality and even won MOLIT's top prize.

Published on: Dec 08, 2025
POSCO E&C debuts AI that automates ready-mix production and predicts strength before 28 days

POSCO E&C launches AI-based ready-mix concrete quality prediction and production automation

POSCO E&C has developed an AI-driven system with manufacturing AI specialist SHLab to produce ready-mix concrete with consistent quality. The platform analyzes mix conditions in real time, predicts outcomes, and automatically tunes production within KS standards.

Anyone running pours knows the problem: even when a batch meets legal specs, variability creeps in through operator skill, material differences, or temperature swings. This approach aims to stabilize slump and strength before concrete leaves the plant-and before it hits your schedule.

How it works

  • Real-time slump estimation: AI analyzes video of the concrete during mixing, infers slump, and automatically adjusts the mix ratio within KS limits.
  • Early strength prediction: The system predicts compressive strength ahead of the usual 28-day wait using mixing conditions and accumulated mix data.
  • Residual water check: It automatically detects and accounts for water remaining in the mixer truck to avoid unexpected dilution and strength loss.
  • End-to-end oversight: POSCO E&C reports a smart management system spanning production, transport, receipt and inspection, placement, and curing.

The technology received the top innovation award at the 2025 Smart Construction Challenge hosted by the Ministry of Land, Infrastructure and Transport (MOLIT).

Why this matters for owners, GCs, and QC teams

  • Tighter variability control: Automated adjustments reduce the risk of off-target slumps and minimize rework or field water additions.
  • Faster decisions: Early strength prediction can inform stripping, lift sequencing, and post-tensioning plans-while still following code and spec requirements.
  • Traceable quality: Digital logs document mix conditions, AI decisions, and adjustments, strengthening QA/QC records and dispute defensibility.
  • Less waste: Preventing strength degradation from residual water protects performance and reduces rejected loads.

Rollout and industry alignment

POSCO E&C plans to build a cooperation framework with public institutions such as Korea Land & Housing Corporation (LH) and major domestic builders. The company will also pursue platform and service businesses around AI-based quality management.

On the policy side, POSCO E&C will work with the government to reflect digital verification for ready-mix production data in official quality management guidelines, and will request applying ready-mix transport information to the Comprehensive Safety and Quality Management Information System for Construction Work (CSI).

What to ask your supplier now

  • Is AI-based slump estimation and auto-adjustment available at your batch plants? Is it locked to KS standards and contract tolerances?
  • Can you provide predicted compressive strength with confidence ranges alongside traditional test reports?
  • Do you track residual water in trucks and log all water additions at the plant and on site?
  • What data exports or APIs are available for our QA/QC system, CDE, or owner handover?

Spec and contract language to consider

  • Require digital verification logs: mix IDs, video-based slump estimates, adjustment setpoints, water addition records, timestamps.
  • Define acceptance criteria for predicted strength vs. laboratory results, including reconciliation and corrective actions.
  • Mandate data retention, access rights, and cybersecurity controls for production and transport records.
  • Set limits on automated adjustments (within KS and project specs) and a clear fallback to manual QC if thresholds are exceeded.

Implementation tips

  • Run a pilot on a single structural element with tight tolerance requirements to compare predicted and tested strengths.
  • Align the inspection and curing plan so field teams know how to use AI predictions without skipping required tests.
  • Train plant and site staff to interpret AI alerts and document decisions in the project QA/QC log.

As digital verification and CSI integration expand, projects that adopt this now will gain better predictability and cleaner handovers. If your team needs a concise foundation in AI for production workflows, consider this practical overview: AI Automation Certification.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)
Advertisement
Stream Watch Guide