Achieve real impact with pre-emptive guidance.
Condals’ three molding lines produce over 43,000 tons of iron castings each year at its two Spanish and Slovakian locations. CIO David de la Cruz stated that the Spanish foundry’s mission “… is to be a data-driven company, taking decisions based on the data from our process …” To this end, the CIO emphasized that in adopting DataProphet’s AI-driven production guidance, “… reducing scrap was always the main goal.”
Together with our strategic partner Norican | DISA (who distributes our solution in the foundry space), DataProphet established a data foundation. Monitizer | PRESCRIBE was the next step to achieving Condols’ scrap reduction objective. During the project, DataProphet PRESCRIBE trained and modeled AI algorithms from live and historical process data on one of Condal’s casting lines. DataProphet PRESCRIBE’s AI-driven analytics were then able to deliver dynamic, real-time process analysis across the entire production line—to significantly and proactively reduce Condals’ scrap for rapid and measurable ROI.
“Other competing solutions only look at parts of the process in detail and only optimize certain parameters – mold, alignment, temperature, porosity – whereas Monitizer | PRESCRIBE optimizes the entire process to reduce scrap. That was exactly what we were looking for.” Condals CIO, David de la Cruz
In collaboration with Condals’ plant personnel, our team extracted, transformed, and loaded Condals’ historical production data into the Monitizer | PRESCRIBE platform. Over 700 parameters described the casting processes—across sand preparation, the melting line, the molding line, and during casting. We used a software tracing technique to ensure the casting process was tracked sufficiently to leverage the industrial edge data.
For Condals, DataProphet's neural network AI model established interactions between hundreds of input processes and machine variables—plus final casting quality data. The procedure specified the optimal operating regime for two of the five patterns under test. Each pattern has a dedicated model generating custom prescriptions to optimize each pattern's recipe. After rigorous checking with test data to confirm the models' predictive accuracy, real-life commissioning began in October 2020.
Time-synchronising the different data feeds in advanced foundries is often challenging. In this case, multiple methods are employed to track molten metal and individual molds and castings—such as mold numbers and pattern keys. Where no tracking method exists, DataProphet typically
introduces software tracing techniques that can detect process events. They include molten metal leaving the holding furnace and being poured. With judicious software tracing, we can calculate the time taken to move between production line events. For example, a spike in pouring ladle temperature and weight will mark exactly when the ladle was filled with metal.
To reduce scrap, MONITIZER | PRESCRIBE delivers pre-emptive process analysis over an entire production line. Commissioning work typically begins in live production. For this foundry client, improvements arrived quickly and easily. As stated by Condals CIO, David de la Cruz: “The great thing about PRESCRIBE is that it brings everything together to predict its influence on scrap and gives you a clearer picture of what is really happening—that is one of its biggest attractions. The prescriptions are pointing us in the right direction, for example, in showing us which variables have the most influence on our process and so what our priorities should be ... Monitizer | PRESCRIBE is also a very easy solution to understand and use. PRESCRIBE is definitely the right approach, it shows us the key ways to reduce the scrap rate and is already giving us excellent feedback."
The scrap rate of one test pattern was already low, but enacting the AI prescriptions reduced it by a further 39%. MONITIZER | PRESCRIBE’s pre-emptive recommendations cut the other pattern’s high scrap rate by 45%. “These initial commissioning results are very encouraging,” explained de la Cruz. “The scrap rate is staying low and actually still dropping slowly ... this is the right approach and we are going in the right direction. We definitely expect further improvements in the future.”
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.