Rethinking Quality: How Artificial Vision Optimizes Pooling Systems

For more than 25 years, the Cartonplast pooling system has been built on a clear principle: consistent quality control.

At Cartonplast, layer pads are manually sorted and inspected after every cycle throughout the process. Trained employees ensure that only pads in flawless condition are returned to circulation.

This system has proven reliable and has formed the backbone of stable and secure supply chains for decades. Yet even a well-established system can be further improved.

With the introduction of an Artificial Vision system, Cartonplast is now taking the next step. At the Leese facility in Germany, an automated, AI-supported quality inspection has been integrated into the existing washing process for the first time. This technology complements manual inspection and opens new possibilities for precision and efficiency. The goal is not to replace a system that already works well, but to make a very good system even better.

Operational Reality in a Pooling System

During their lifecycle in glass and can logistics, layer pads are exposed to significant mechanical stress. They carry weight, stabilize stacks, are washed, dried, transported and used again. Every cycle places stress on the material. Typical challenges are therefore not exceptions but part of everyday operations.

Damage and weak points are often difficult to detect with the naked eye. At the same time, heavy loads can cause deformation that may lead to instability in the next cycle. Contamination is another factor. It is not always immediately visible but can still be relevant from a hygiene perspective.

For this reason, manual visual inspection remains a central element of quality control. It relies on experience, trained observation and clearly defined quality standards. However, as circulation volumes increase and pooling systems continue to grow, an additional question arises: how can this proven control mechanism be complemented by data-driven technologies?

AI-Supported Inspection as an Additional Quality Gate

At the Leese facility, an automated visual inspection has been integrated for the first time, digitally capturing each layer pad during the washing process. High-resolution cameras analyse every pad in real time.

Algorithms systematically check for:

holes, missing sections, cracks, contamination and deformation.

The evaluation is objective and reproducible. Every pad is assessed according to the same criteria, regardless of time pressure or subjective perception.

Importantly, technology is seamlessly integrated into the existing washing process. Inspection takes place in-line, without delays. Pads are automatically sorted and removed from circulation when necessary.

What Artificial Vision Changes in Practice

What does this technological enhancement mean for day-to-day pooling operations? Not in theory. Not on PowerPoint slides. But in the real environment where layer pads pass through washing systems and circulation loops thousands of times every day.

When quality decisions are supported by data, four key performance dimensions of a pooling system improve.

1. More Precise Lifecycle Management

Early detection of damage allows pads to be monitored before failures occur. At the same time, functional pads are not removed from circulation prematurely.

Within a Total Cost of Ownership perspective, this results in:

more stable circulation numbers

more predictable replacement requirements

2. Improved Resource Efficiency

Every additional circulation cycle increases the resource efficiency of a layer pad. When pads are used for exactly as long as their technical condition allows, the need for replacement material decreases.

Material consumption over the entire lifecycle is reduced. At the same time, the CO₂ footprint per use improves, as the production effort is distributed across a higher number of cycles.

3. Even More Stable Transport Performance

A pooling system with consistently verified quality behaves more reliably. When damaged or deformed pads are detected early, the risk of unstable pallet stacks decreases.

This reduces:

transport damage

product losses

rework at production sites

complaints along the supply chain

In sensitive industries such as food, beverage or pharmaceuticals, this additional level of security is a significant advantage.

4. New Transparency Within the System

Artificial Vision also creates a comprehensive data foundation on the condition of deployed assets.

This generates deeper insights into:

actual circulation cycles

typical damage patterns

lifecycle development of individual assets

This transparency supports not only operational management but also reporting and compliance requirements.

Innovation as the Next Evolution of the Pooling System

The introduction of Artificial Vision at the Leese facility represents an important step in the ongoing development of the Cartonplast system. Manual quality inspections remain a core component of processes across all facilities. At the same time, the integration of intelligent technologies demonstrates how an established system can be further optimized.

Cartonplast combines decades of operational experience with advanced technology, setting a new benchmark for quality and transparency in pooling systems. For partners, this ultimately means one thing: collaboration at the highest quality level — today and in the future.