Causal AI revolutionises manufacturing efficiency and quality control

Ethon.ai is making waves in the use of AI in manufacturing and promises to revolutionise industries with its innovative approach. Ethon.ai provides a software application that leverages artificial intelligence to unravel the complexities of manufacturing processes. Recently recognized by Forbes as one of the rising stars in the software domain, this application is poised to transform how we understand and optimise production lines.

At the heart of this technological marvel is the concept of Causal AI, a term that might not be familiar to everyone but holds immense potential in the realm of manufacturing. While the buzz around generative AI has captured the imagination of many, Causal AI offers a similarly transformative promise. It focuses on establishing cause-and-effect relationships within data, enabling manufacturers to simulate and predict outcomes based on various parameters.

Imagine a bustling factory floor, teeming with machines and sensors capturing data on temperatures, pressures, and more. Causal AI, the driving force behind this application, seeks to decode the intricate web of relationships between these parameters. By doing so, it empowers manufacturers to simulate changes and optimise key performance indicators, ultimately enhancing product quality.

The process unfolds in three distinct steps. Initially, the AI learns a causal graph by analysing data and mimicking the physical relationships between parameters. This graph serves as a blueprint, illustrating how different factors interconnect. Next, the AI estimates the magnitude of these effects, providing insights into the impact of each parameter. Finally, simulations are conducted to explore "what-if" scenarios, allowing manufacturers to anticipate the consequences of altering specific variables.

To illustrate the power of this technology, consider the case of chocolate manufacturing — a universally beloved industry. A renowned chocolate producer, stands as one of the application's largest customers. The manufacturing process involves preheating moulds, filling them with chocolate, cooling, and quality control. Traditionally, quality control relied on computer vision systems to detect defects, but Causal AI takes it a step further.

Picture a process engineer observing a quality curve over time, only to witness a sudden decline. Chocolate breaks, sticks, and ultimately goes to waste. Identifying the root cause of this waste is crucial, not only for environmental reasons but also for cost savings. In the past, engineers relied on gut feelings and rudimentary statistical analysis, often leading to inconclusive results. However, with Causal AI, a digital twin of the factory is created, mapping each process parameter to its respective step.

By feeding data from the past month into the AI, it uncovers hidden relationships within the manufacturing process. Parameters such as temperatures, pressures, and equipment migration are analysed alongside quality data, resulting in a causal graph. This graph reveals five distinct branches influencing quality, each representing a potential root cause.

Delving deeper, the AI identifies mould temperature as a critical parameter affecting quality. An inverse U-shaped relationship emerges, indicating that deviations from an optimal temperature range lead to quality losses. While mould temperature itself may not be directly adjustable, the AI traces the Causal graph back to heating power—a parameter that can be manipulated.

?Armed with this knowledge, manufacturers can explore how changes in heating power impact quality. Simulations conducted by the AI estimate the optimal setting, pinpointing 865 watts as the sweet spot. This insight allows manufacturers to visualise the potential quality improvements they could have achieved in the past, had they operated at this optimal setting.

?The implications are profound. By harnessing the power of causal AI, manufacturers can prevent losses, reduce waste, and save millions annually. This transformative technology not only enhances efficiency but also contributes to sustainability efforts by minimizing resource wastage.

In a world where data reigns supreme, Causal AI emerges as a beacon of innovation, offering a glimpse into a future where manufacturing processes are optimized with precision and foresight. As industries continue to embrace this technology, the potential for positive change is boundless, promising a brighter and more efficient tomorrow.