Factory Digitalization_Ver01. _ "The Benefits and Limitations of Digital Transformation in Sand Casting Foundries" - What Should Be Adopted Together?

Mar 03, 2026

Introduction

Hello everyone. Today I'd like to talk about a topic that has become increasingly popular in the foundry industry: IoT, digital transformation, visualization, and AI-driven optimization—all from a practical, shop-floor perspective.

In my previous role at a global manufacturer of automatic molding machines, sand plant systems, and shot blast equipment for sand casting foundries, I led multiple digitalization and AI implementation projects. Based on that hands-on experience, this article summarizes the benefits and limitations of digital transformation in sand casting operations.

Sand Casting Is a Historically Established Industry

Casting technology was introduced to Japan during the Yayoi period (around the 3rd century B.C.), and sand molding has existed since that time. Today, sand casting is used across automotive, construction, and industrial machinery sectors. Many everyday products—such as manhole covers and cooking pots—are also produced by casting.

Modern foundries, however, face a very different reality. The production flow— sand preparation → melting → inoculation → molding → pouring → cooling → shakeout → shot blasting— is highly interconnected, and one process is almost always a bottleneck.

This leads to structural challenges:

• Frequent waiting and downtime

• Difficulty increasing throughput

• Complex root-cause analysis for defects

• Limited traceability across processes

These issues are exactly why digital transformation is needed.

Digital Benefit #1: Real-Time Visualization

A System to Collect and Display Data

The first requirement is real-time visualization of all production data.

• Collect raw data from PLCs and PCs

• Upload to a central database

• Display instantly on shop-floor monitors or client PCs

To achieve this, foundries need:

• A scalable cloud platform (e.g., AWS)

• Encrypted gateways for secure data transfer

• A secure factory network infrastructure

User-friendliness of Dashboards Matters

Visualized data should be displayed in dashboards with flexible widgets:

• Bar charts

• Line charts

• Histograms

• Table Widgets

Needless to say, all data must be exportable as CSV files.

Digital Benefit #2: Bottleneck Analysis

Embedding Logic for Productivity Metrics

Visualization alone is not enough. To improve productivity, the system must analyze:

• Waiting time per root cause

• Downtime per root cause

• Throughput

This requires embedding calculation logic directly into the dashboard.

Digital Benefit #3: Unified Manufacturing View (Traceability Foundation)

Linking All Processes to a Single Pouring ID

To correlate defects with manufacturing processes, all process data must be linked at the smallest production unit (Batch or Lot).

This includes:

• Pouring ID

• Melting data

• Inoculation data

• Material charging history

• Molding ID

• Sand mixing batch ID

Once these are connected, the full chain becomes visible:

Defect → Pouring → Molding → Sand preparation → Melting → Inoculation

This is the foundation of true traceability.

Digital Benefit #4: AI for Multi-Dimensional Defect Reduction

AI Can Analyze What Humans Cannot

Casting defects often arise from multiple interacting factors. Human analysis is typically limited to two dimensions, but AI can analyze multi-dimensional relationships and propose actionable adjustments.

Examples:

• Adjusting sand moisture

• Fine-tuning pouring temperature or cycle

• Optimizing in-mold cooling time

In one of my projects, AI reduced defect rates by more than 80%, even after the customer had exhausted all conventional improvement methods.

Limitations of AI

Limitation #1: AI Depends on Each Foundry's Historical Data

Because every foundry has different equipment, materials, and environmental conditions, a universal “ideal model” cannot be applied. To expand AI's learning range, foundries must intentionally introduce textbook-like process conditions as test runs.

Limitation #2: AI Cannot Address Design-Driven Defects

Defects caused by:

• Pattern design

• Core design

• Gating and riser layout

cannot be solved by AI. These require MAGMA simulation or consultation with application experts.

Conclusion

AI is not a magic solution, but it is a powerful tool for process improvement. At the same time, pattern design, core design, and gating design remain areas where human expertise is essential.

Digital transformation is not the goal—it is a means to stabilize quality and cost.

Contact

If you are considering factory data visualization or struggling with persistent defect rates in specific products, feel free to reach out.

60-minute free initial consultation info@metricjapan.com

We look forward to supporting your improvement journey.

Tags

#sandcasting #sandmolding #ironfoundry #digitaltransformation #IoT #datavisualization #AI #defectreduction #manufacturingDX #productivityimprovement #traceability #unifiedmanufacturingdata #moldingmachine #sandpreparation #pouringprocess #qualitycontrol #MAGMA #processconsulting #applicationconsulting #manufacturingconsulting #factoryimprovement #costoptimization #manufacturingdataanalysis #MetricJapan