AI for Foundries

Learn how Artificial Intelligence (AI) helped a major engine block manufacturer attain zero percent external scrap.




Find us at the 2019 Foundry Leadership Summit


Your Message

“I have been using your prescriptive reports to make small calculated adjustments to our system. The norm for us is 5-6% internal scrap and 10-15% rework. Over a two day period we managed to achieve 1% internal scrap and 8% rework. In the past we have been able to achieve similar results, but we had absolutely no clue what we did to achieve the good result. This time around we have a fairly good idea of what we did to achieve the result. Thanks for your tremendous efforts and help to take us to the next level.”
Chief Executive Officer

Improve output in foundries with AI

The world is evolving ever more rapidly and the foundry industry is no exception, data is at the heart of the fourth industrial revolution. Often stored in disparate sources across a plant, timely analysis and actionable insights are key to unlocking the immense value encoded in plant measurements. The application of these prescriptive actions is used to achieve the optimal state of production and improve output in foundries. Artificial Intelligence (AI) is the innovation that makes this possible.

Through numerous implementations of our AI solution, DataProphet has provided clients with expert advice proven to deliver actionable and measurable results for continuous improvement in plant operations. This is the realisation of the Industry 4.0 vision in foundries.

DataProphet’s team of highly skilled engineers and data scientists partner with clients to gain a thorough understanding of the processes within the plant, identifying relevant data sources and mapping the data environment in the context of the physical plant. DataProphet builds a unified view, linking both quality and process data. Upon ingestion of the data, our AI solution identifies an optimum operating paradigm to reduce variation and prescribes actions to achieve the optimal yield, resulting in more predictable plant performance.
Total Business Costs Saved with no process changes
External Scrap
4 - 8
Week Deployment
1 - 5 Years
Historical Data Analyzed
Production Increase
Product Traceability
>700M records
5-10k Features
In a single view

DataProphet's Foundry Solutions

Using only the production process data, DataProphet PRESCRIBE can predict the most likely defect locations for given process variable values. The defects are always reported per physical location on the castings.

DataProphet PRESCRIBE also identifies a desirable operating region and generates an understanding of how key process variables differ as a function of yield regions. This would show clear progression across the Principal Component Analysis (PCA) space over time, with surface visualization of good vs. bad for key variables.

Using region based convolutional neural networks upon finished products, improve quality control (QC) inspection tasks by looking for subsurface defects that occur most commonly in the client's products.

Create a unified view of production data

Manufacturing Case Study

Predicting engine block defects and identifying high yield operating regions.

Product Sheet

$400,000 monthly savings and 0% defects on shipments.

Computer Vision Case Study

Machine vision systems for enhanced quality control.


Typical Foundry Process Flow