Achieve real impact with pre-emptive guidance.
Our client is one of the largest foundries in the Southern Hemisphere. Each year it produces 129,000 high-quality cast iron automotive components, or over 46,000 tonnes, at its plant. The plant has an annual melting capacity of 110,000 tonnes. In order to compete globally and to achieve its vision to be the best foundry in the world, our client embraced the opportunities of Industry 4.0. As a result, it has become one of the most innovative and advanced foundries in the world.
They “embraced the information flow and sharing principles that underscore what is commonly referred to as Industry 4.0,” our client explained. “We upgraded and increased digitization and the acquisition of information across all stages of production. A natural outcome of this strategy was to use the production and quality data to enable plant engineers and managers to make better, more informed decisions.”
Our client wanted to find a way to optimize its manufacturing process to reduce the cost of scrap. While the scrap rate was no higher than the industry standard, every defective unit shipped and scrapped in Europe or North America incurred a relatively high shipping cost.
The main objective was therefore to lower the number of defective engine blocks shipped, by reducing the internal and external defect and scrap rates. In turn, this would increase production, and reduce rework.
This is not a novel objective for many plant managers and process engineers across the world. The challenge, historically, is in the level of complexity in the large number of non-linear causal relationships that make up the modern foundry.
DataProphet PRESCRIBE optimizes production at the client by providing continuous prescriptions to operators across the entire manufacturing process. The client uses DataProphet’s prescriptive setpoints to operate the plant within, or very near the identified optimum region, allowing for a minor degree of variability due to non-controllable process parameters. The prescriptions are updated every five minutes based on data that is continuously ingested from the process.
DataProphet simultaneously implemented the extraction, transformation, and loading (ETL), as well as the warehousing, of data from across the entire plant. Historical production data was gathered from stakeholders across multiple departments, from PLCs and from the plant’s central SCADA systems. The data existed in a number of different formats, including handwritten forms, excel files, proprietary databases, and CSV files. DataProphet then digitally transformed this process history into a single view spanning 15 months of historic production data, 173,000 records, and 400 unique process variables. The extraction of data from these disparate sources was as important as its analysis.
Once digitized, DataProphet fed the process and quality data into the DataProphet PRESCRIBE system. Through the application of advanced supervised and unsupervised machine learning methods, DataProphet PRESCRIBE automatically discovers the optimal operating regime for the client’s complex, multi-step, industrial process. PRESCRIBE uses a deep, monolithic model of the process, which encapsulates the correlations and interactions between approximately 1,000 parameters from across the plant.
Too often, AI-as-a-Service providers equate research-based methodologies with one-size-fits-all industry solutions. DataProphet works closely with partners to gain an understanding of their unique set of production problems. We adopt AI-driven manufacturing solutions to the production dynamics of your factory floor, incorporating our proven, industrial process expertise, but taking our cues from contextualized feedback. We understand that most impact is achieved by building machine-learning models to mesh with your production lines' realities and the assets that drive them. Our partnerships are guaranteed win-win.
Sign up to receive personalized, hot-of-the-press news about DataProphet events, and our latest innovations and products.