Why Machine Learning matters for the perfect performance of RFID tunnels

Using RFID makes handling processes in warehouses and distribution centers more efficient by leveraging the technology's capabilities to capture products in bulk, distance, and even in closed boxes. As a result, product data is captured seamlessly, efficiently, and up to 25 times faster than manual handling.

With the help of RFID tunnels, it is possible to automatically receive hundreds of boxes per hour and know every single item inside, even down to its size and color. This enables brands, distributors, wholesalers, and other logistic parties to identify under-deliveries or other errors to ensure that shipments are accurate and the distribution center can operate based on correct inventory information.

Streamlining product registration with RFID tunnels?

If RFID tunnels are employed for seamless shipment verifications, they must function with extreme accuracy. Otherwise, you'll have a drop in efficiency (manually having to verify a lot of boxes after the tunnel) or wrongfully blame suppliers for not shipping the right quantities.

In practice, the RFID reader must read everything inside the box but nothing of the box before and after - and any packages that might linger around the tunnel. Especially at high speed, low separation, and larger box quantities, this becomes a significant challenge.

The easy part is reading everything you want to read. You can use as much RFID power as possible, but should also consider the efficient configuration of how the RFID reader talks with the RFID tags. This should be optimized for speed, but there is a danger of suffering interference from other readers. Therefore, mitigation techniques should be applied.?

Making sure to read the right things?

The challenge is to avoid so-called “false-positive-reads”: RFID tags from the previous and next boxes in line. The first step is to select the best RFID antennas for the job that read very carefully close by but not too much farther away.

Also, the placement of the antennas is of critical importance. However, this will not solve all stray tag reads. Some RFID tunnel devices use mechanical metallic doors or roller shutters in front and after the tunnel to shield the box you want to read from boxes before and afterward. However, this solution is most likely to slow down the throughput, is more expensive, and places an additional burden on maintenance due to the mechanical components involved.

The better way is to use Machine Learning! In the past few years, Machine Learning has come a long way - most recent examples include the ChatGPT chatbot, which offers an almost human-like conversation.?

Using Machine Learning in RFID tunnels??

With machine learning for RFID tunnels, it is possible to calculate for every RFID tag read by the reader how likely it is that it was included in the box that we want to read or whether it was a 'stray' read. This means an RFID tag belongs to a box before or after the one you want to read - or a box just lying around the tunnel.

The algorithm's accuracy needs to be extremely high: if you have 50 items in a box and want to estimate the box's contents correctly, the algorithm needs to be right 50 times. So, having a 99.9% correct algorithm will still yield one in thousands of estimates to be wrong, and thus a one-in-twenty boxes false rejects. The algorithm, thus, needs to be very accurate. Therefore more than thirty parameters have been carefully selected for the algorithm when deciding on the accurate read of an RFID tag.

In addition, the algorithm was trained by reading thousands of known boxes of different vendors and brands in different reading set-ups. This has made the algorithm robust enough to run at high accuracy at high speed - in some circumstances, only with a simple curtain at the tunnel's entrance or no shielding at all.?

One RFID platform??

TunnelML is part of the Nedap iD Cloud platform. With Nedap's TunnelML algorithm, it is possible to design and use RFID tunnels that require less hardware.

For example, the algorithm can accurately read and identify the contents of boxes without mechanical metallic doors or roller shutters, which can slow down the process and increase maintenance costs.

As a result, initial investments will be lower, and the total cost of ownership will be reduced because there is less hardware to maintain, which can fail.??