September 17 2018
There is currently a huge wave of RFID adoption going on in fashion and apparel retail. This growth is largely based on a business case that can be easily made: the need for accurate stock data is critical for today’s retailers; while at the same time the operating cost for deploying RFID has significantly decreased due to lower RFID tag prices that are now below three eurocents a piece.
Considering that the price was at ten eurocents only five years ago, this has made a big difference and significantly improved the business case. An even lower price would allow for adoption in other markets, like food retail. The question is now: is there any more improvement to be expected or has the bottom been reached?
Current price structure
To answer this question, we need to understand how the price of an RFID tag is currently set up. The price of an RFID tag consists of the following core components:
- The antenna plus substrate
- The RFID chip
- Attaching the chip on the antenna
- Margin for all of the companies involved in the process
(We don’t consider label conversion, encoding and printing for now).
New ways of creating the antenna
The RFID antenna (plus the substrate it is attached on), can be created in different ways. Traditionally, an ‘etching’ process is used to make those antennas. Etching works by creating an aluminium layer on a plastic film - and then etching away the parts that are not needed for the antenna with chemicals. The aluminium that is etched away is wasted and the chemicals used for this process can be hazardous for the environment.
Newer methods involve printing the antenna using conductive ink, cutting the antenna with a rotating drum or laser cutting the antenna. All those methods are less wasteful, and allow the antenna to be directly attached to paper - doing away with the plastic film that is required for the etching process. This saves costs and significantly reduces the environmental footprint for RFID.
Another advantage of the new antenna creation methods, is that those processes can be ‘in-line’ with the chip attachment process. With etching, creating the antenna is typically done in another environment (and typically by a different company, than the company attaching the chip). If the antenna can be created by the same company, in the same factory as the chip-attach, this saves costs and logistic complexities - another opportunity for cost price reduction.
Direct Die Attach - Picture courtesy of Muhlbauer, “Concept 2020 - Roadmap for high volume RAIN tag manufacturing”
Smaller chips lead to lower prices
The price of the chip is heavily related to size of the chip. However, the smaller you make the chip, the more difficult it is to handle and manage the chip. Traditionally, RFID machines where based on a pick-and-place technology that was similar to how Printed Circuit Boards were produced. The disadvantage of this process, is that it requires a lot of moving parts and that it doesn’t allow for very small chips. According to a presentation of chip-attach machine manufacturer Mühlbauer, they have a new way of working which is called ‘direct-die-attach’. Direct-die-attach moves the wafer over a tape of RFID antennas, and the chip is directly attached to the antenna when it is ejected from the wafer.
With this method, it not only allows for much higher throughput (40.000 units per hour in the current machine), it also allows for smaller RFID chips. The current size of an RFID chip is approximately 0.5 x 0.5 mm - and of course size is an important driver for chip cost. It is expected that chip manufactures use this potential for size reduction by switching to smaller chip sizes. This is standard practise in manufacturing chips: by using more advanced manufacturing processes, the size of components can be decreased.
A 50% reduction in storage on a chip could give even lower prices
Another variable at play is the amount of available memory on tag chips. Current tag chips still have memory that stays unused in retail applications. 96 bits are needed to store the product information and serial (the SGTIN) - but there is a lot more memory available on the chip. This memory is for example used for:
- The TID serial: a unique serial number that is programmed at chip manufacturing, and globally unique. As we already have the serial number in the EPC, it is not strictly necessary to also have a serial number in the TID. However, because the serial number is programmed at chip production, it is guaranteed to be unique. This proves valuable in detecting EPC duplicates or product authentication checks – but that is not always relevant.
- The kill password: The kill password is only relevant when you need to kill tags. In most situations, this is not strictly necessary. The access password however is still relevant - mainly to control privacy features like read range reduction.
- Additional EPC memory (or user memory): Some ICs have more EPC memory (or user memory) than the required 96 bits for retail applications. This is most often not used in retail applications, as additional information is stored in a cloud-system, where storage is much cheaper.
If the kill password, the TID serial and the additional EPC memory are removed from RFID chips, this can save up to 50% of the storage of the chip, whichwill again reduce the price.
Conclusion: the cost can be reduced at all components
To summarise, for each of the cost components in an RFID tag there are optimisations available:
- The antenna plus substrate: newer manufacturing methods lead to less waste and less material required, leading to lower cost
- The RFID chip: more advanced production technologies allow smaller chips, and additional savings can be expected in reducing the memory footprint of those chips
- Attaching the chip on the antenna: new manufacturing technologies like direct-die-attach allow for higher throughput, and lower per-unit cost
- Margin for all of the companies involved in the process: integrating the antenna production with the chip-attach in one machine and facility simplifies the supply chain
Whether this will give us the one cent RFID tag remains to be seen, but it at least shows that there is significant room for improvement over the next few years. If the price drops further, it will allow spreading RFID to other markets than just fashion/apparel. Potentially even stretch into food retail, opening up all kinds of new use cases.
Product Developer
Danny Haak
