Our customer Ingredion wanted to optimize its plant capacity through data-driven predictions.
Reaction times in production were controlled by employees on the basis of experience, in-process laboratory analyses and Excel data.
The hypothesis:
Ingredion, a leading company in the food industry, was about to embark on an innovative process. For years, Ingredion relied on the experience of its plant operators, extensive laboratory analyses and conventional Excel spreadsheets to control production processes. But in an increasingly complex domain that is constantly evolving, it was time to go one step further – towards data-driven solutions. In this case, an assistance system that predicts the optimal response times.
The challenge: optimization through data
Ingredion faced a key challenge: how could the company maximize its plant capacity while minimizing energy consumption? The traditional methods were good, but not good enough. Manual estimates tended to play it safe, which probably led to less efficient operations.
This is where our team came into play. Our joint hypothesis was simple but innovative: with accurate AI-driven predictions, we could help Ingredion increase plant capacity and reduce energy consumption. We developed an advanced AI model that was able to predict production processes more accurately and faster than ever before.
The breakthrough: measurable results
The results spoke for themselves: our predictions in the MVP were on average 5 to 10 minutes more accurate than the plant operators’ estimates. This led to an impressive increase in system throughput of 5-10%. This optimization meant several million euros in additional net profit per year when rolled out to all plants. Even more impressive was the fact that the prediction error of our model was within the measurement tolerance of the samples – in other words, it could hardly get any better.
This transformation marks the beginning of a data-driven strategy at Ingredion – in cooperation with Ailio, data and AI are the driving forces behind more efficient, profitable and sustainable production processes. As a partner of Ingredion, we are proud to accompany this journey and look forward to pushing the boundaries of what is possible with data science.