RFID Deployment Faces Cost Hurdle

A report from high-tech consultants {complink 9171|Frost & Sullivan} claims that the use of Radio Frequency Identification (RFID), the tracking technology used by many electronics manufacturers, will grow in the Asia/Pacific region this year.

Frost and Sullivan’s analysts cite increased use of the technology by supply chain managers “in Asia Pacific countries that are suppliers to big retailers in the United States,” as a driver of the growth. The argument for RFID is, they claim, efficiency. Tagging items at their source, rather than at their destination, should save time and streamline delivery. Or at least that’s Frost and Sullivan’s take:

“With goods being tagged at the manufacturing or packaging stage rather than the store, store associates are left free to focus on customer service. Source tagging thus helps enhance sales and reduces shrinkage in the supply chain.”

One problem: it’s not cheap. For a large manufacturer, installing an RFID system through an entire supply chain will cost, by Frost and Sullivan’s estimates, between $10 million and $25 million. Most of that expense is in hardware. So-called “source tagging” — installing the RFID strip, usually a thin strip of magnetic material, at the production or packaging plant, rather than at the point of sale — requires installing detectors at every point in the supply chain.

Though familiar to most people as an anti-shoplifting technology, it’s more often a technology for stopping inside jobs. As summed up by the Discovery Channel’s How Stuff Works :

“These invisible tags, which are deactivated by the clerk with a verifier that needs no physical contact with the tag to work, are especially effective at addressing employee theft and represent a hot topic in retail security today.”

In 2009, RFID in just the Asia/Pacific region was a $600 million to $800 million market, claims Frost and Sullivan. So for the market to be growing, manufacturers have to believe that by employing RFID tech, they will save at least that much. They’re proposing to stop eight-figure losses over the life of the equipment.

It’s a big bet. RFID’s advantage is its suitability for use throughout a supply chain, where older, balkier microwave systems are more appropriate for use at points of sale. But better technology may not be enough. Older supply chain security systems have the advantage of already being bought and paid for. Even if they leak a bit, they may not leak to the tune of $25 million, or enough to make them worth replacing. All those knock-off iPads sold in illegal electronics bazaars may be painful for electronics manufacturers. But accepting current levels of supply line leakage may be easier to swallow than the bill for RFID.

Of course, the real cost depends on the value of what’s being stolen. Growth outside the electronic sector has been strong for the technology. It’s also showing up in surprising places. The same week Frost and Sullivan released its report, Singapore’s Marina Bay casino reported that it would be installing RFID on its gaming floor — chips in the chips — after a group of fraudulent card players tried to pass off fake $1000 ducats as winnings.

If an RFID tag can keep someone honest in a casino, it’s reasonable to imagine it keeping pen drives, digital camera bodies, and circuit boards secure in an international supply line. But to install $25 million in gear also means admitting, to your supply partners and your investors, that you’re legitimately worried about $25 million in thefts. And that has a cost, though harder to calculate, too.

33 comments on “RFID Deployment Faces Cost Hurdle

  1. Eldredge
    January 24, 2011

    I would think another benefit of RFID technology would be the ability to track legitimate merchandise through the distribution system. and help reduce the likelihood of knock-off merchandise.


  2. Eldredge
    January 24, 2011

    Another factor that may need to be considered in the cost of introduction of RFID tags: how easy is it to cheat the system, and therefore continue losses in spite of the investment? RFID tags are not as invisible as one might like. I'm sure some clever people will find ways to beat them, and the value of investing in the RFID system goes down as their success rate goes up.

  3. Marc Herman
    January 24, 2011

    It does seem like a big, beta v. vhs-type bet, insofar as if people do figure out how to end-run RFID, then you've spent $25 million on something that doesn't work. If the Frost prediction is right, then enough Asian companies are confident enough in the tech to believe it's worth the investment. People are wrong all the time, of course. Thanks for your comment.

  4. seel225
    January 24, 2011

    I admit that RFID deployement faces cost hurdle but it makes our lives easier, Imagine going to a grocery store or any electronic store during thanksgiving time, Instead of waiting in the line for an hours, filling up your cart and walking out the door, wow….It saves lot of time. It costs so much money to setup the system but for big organizations i don't think its a big deal.

  5. Eldredge
    January 24, 2011

    I'm sure that system developers have spent a significant amount of due dilligence in trying to make the systems foolproof, and it is bound be be an improvement over current methods. RFID has been in development and testing for a long time.

  6. Barbara Jorgensen
    January 24, 2011

    For the electronics supply chain, it certainly make sense to tag items at the source–and cost is the main reason the technology is going to remain in limbo. For high-priced components, suppliers would tag them at the individual component level; others could be tagged at the lot or even pallet level. But even at pennies per tag, the cost is overwhelming for component-level suppliers. It would even add up at the subassembly level. It's too bad: not only would RFID avoid theft, it could help authenticate parts and prevent counterfeiting. 

  7. Marc Herman
    January 24, 2011

    Yeah. It doesn't seem likely to pencil out. So then the mystery is: why is Frost, which is the sort of place rigorous enough to ask, and get, $10,000 a copy for some of its research reports, predicting use, and sales of RFID systems, to rise in Asia this year, including the electronics sector? It's making sense for someone. Just not anyone I can find speaking publically about it.

  8. SP
    January 24, 2011

    I am not sure if I understoood correctly can deploing RFID cost 25 million dollars?? Is it so expensive. One of my friend and former colleague started his own company that focusses on developing RFID products. Well I need to do some cost check…

  9. jbond
    January 24, 2011

    I think RFID is a very valid and useful tool with multiple uses. Most analysts and companies agree with the effectiveness of this great tool. The only downfall everybody faces the large cost.

    This is a much needed technology that would help the entire industry by not only helping to slow down the black market sales, but also allowing companies to get a better control of theft. With the ever expanding market in Asia in not only production but purchasing, there will be some manufacturers willing to take the cost on to help the company overall.


  10. Eldredge
    January 24, 2011

    I tend to agree that RFID would benefit the black market impact.

  11. elctrnx_lyf
    January 25, 2011

    If a casino can employ the technology, I mean it is real money. They are ready to invest in making chips inside the chips. The electronic manufacturers will definitely begin deploying the RFID in the coming future. If it can provide the benefits of tracking a device from start to sell defnitely it is welocme to be used.

  12. Eldredge
    January 25, 2011

    I agree – the ability to track specific units all the way through the distribution system to point-of-sale without human intervention will be a very powerful tool. I do think the casino model is a little different implementation and cost model than distribution (although an excellent use for the technology) because I would expect the RFID equipment and the chips to all remain inside the grounds of the casino, and the chips will be infinitely recycleable within the faciility. By contrast, the RFID tags or chips in distribution will probably be used one time, however the equipment in the distribution system will be tracking a huge volume of products and quantities and therfore, provide greater efficiency for the equipment than the casino implementation. Both are great uses for the technology.

  13. efficiency
    January 25, 2011

    RFID is not new and was design just forr this purpose.

    The cost here is exploded at least an order of magnitude.

    (marketers or journalist love to throw a few extra zeros on the back end of #'s)

    The supply chain in the middle (shippers) if it is a real supply chain would allready have bought the equipment and it get ammortorized over many customers and products.

    Not every spot along the supply chain needs every rfid tag checked or not at all.

    Just the shipping container would need one rfid tag only, not every tag within the container.

  14. efficiency
    January 25, 2011

    why not the same thing in IC's pt the RFID tag in expensive IC's

  15. Marc Herman
    January 25, 2011

    The 10-25 million stat comes from the report cited in the story. It's certainly useful to ask if those numbers were somehow inflated in the original research. If they are indeed “throwing an extra zero,” the question I'd ask is why? I'm not sure who benefits from overstating the costs; it seems far more likely for a consultant like Frost, or a maker of RFID systems, to lowball the costs by five or ten percent. It would be helpful to see some examples of the cost exaggeration you're saying you encounter, and help us understand what's behind that. At the moment I'm more apt to believe that there really are prohibitive deployment costs, but that those aren't enough to limit the market. It's not a new technology, as you say, and it's a very useful one. But it does seem like there's still some indecision out there about its use. That's why it seemed to me worth noting a predicted increase in RFID for Asian supply chains. Thanks for your comment.


  16. efficiency
    January 25, 2011

    So why and how can walmart use RFID tags on its pallets of toilet paper then?

    Are they not paper thin margin products? (sorry could not help the punn)

    If the low price leader is using them from china to the US stores and it pencils out for them why not for everybody?

  17. Marc Herman
    January 25, 2011

    I'm not sure comparing the profit margins of different industries — in this case paper products and electronics — is quite the point. Here's what we know: some industries have adopted RFID broadly. Other's haven't. A study that appears to be taken seriously by the  business information market, at least as indicated by the premium its authors are charging to read it, suggests that RFID use could increase in Asia this year. But, the researchers claim, costs are a barrier. So what's happening?

    I have no idea what the markup is on bogroll. I also don't know what a supply chain looks like for a company like Kimberly-Clark (inventor of the paper towel, just to pick something). I would suspect toilet paper has fewer steps in its manufacturing process than does, say, a wide-screen TV. So I can imagine different supply chain costs for the products, and different hardware demands were one to decide to install RFID in each chain.

    All that said, you sound like someone deeply involved in this issue. I'd love some specific examples from the electronics industry, such as you have them at hand. Thanks again for commenting.



  18. efficiency
    January 25, 2011

    Just the point that if a paper thin margin biz like walwart can do it and make economic sense so should electronic suppliers..

    (They sell electronics too)

  19. Hardcore
    January 25, 2011

    Casinos do it for different reasons, also the model is completely different.

    1. Security and faking of ' casino' token is the reason they do it.

    2. The cost is not 'written off' per product, the 'token' is re-used multiple times, passing back through the  croupier and the  exchange point when  'real' money is returned to the player in exchange for the  token, making it a highly cost effective system.

    Also EACH PRODUCT needs an RFID chip rather than just the packing case of multiple items, since once the outer packing is stripped off, the inner items would become un-traceable, making the whole system pointless.

    The Idea is that each item is traceable from source to destination (the checkout- point and beyond)


    The main issues related to 'beyond' is that it makes you personally traceable, and also the value of any items tagged in an RFID way, as you walk from store to store in a shopping complex then each shop can see what you purchased in other stores and the value, they can then target you for advertising.

    The same way google can track me to 20 meters, without a mobile phone/ GPS system, because they recorded all the WIFI systems when they did their street view.


  20. Eldredge
    January 25, 2011

    I could be wrong, but I suspect Walmarts' implementation doesn't entail RFID tagging from the source, as discussed in the article, but probably starts once pallets of merchandise enter their distibution system. I'm sure that's an expensive proposition as well, but it sounds like tagging from the source is an order of magnitude more specific in the supply chain, and probably a bit more expensive per unit of production to put in place.

  21. efficiency
    January 26, 2011

    If  you just moved the point where rfid tags are placed to the source of production no added costs would be incurred. Only if points in between are added for tracking would costs go up. The readers are portable and only need to buy as many as you have people operating them. or they are placed at entry/exit points to check in and out of inventory thus only need to implement at those points. This saves valuable human time and this is where the savings come from add theft reduction and that is just icing on the cake.

    I suppose the labor cost in china do not offset the capital as fast as developed countries. But it always pays to keep your people doing the highest value tasks as possible offloading tasks to computers allows this re-tasking = higher productivity.


  22. efficiency
    January 26, 2011

    I guess some of the costs shift across company lines here.

    A multi company approach is what is needed that shares the cost if it is truly as high as the article claims. I am still skeptical.

  23. Taimoor Zubar
    January 26, 2011

    While preventing theft of inventory is a big advantage of RFID tags, tagging inventory in factories offers the advantage of tracking the inventory while it's moving through the supply chain on the way to stores. Also, use of RFID in retail stores makes the checkout process much quicker since RFID tags can be read at a distance. I think cost savings arising from these factors should also be considered.

  24. efficiency
    January 26, 2011

    yes I agree the cost to implement should be weighed against all cost savings.

    The rub is that the cost and saving can cross company lines.

  25. Eldredge
    January 26, 2011

    OK – But I guess I'm making the assumption that tagging is changing from a pallet level to a production unit level….which would provide significant benefits analyzing movement through the distribution chain to the comsumer, but increase the quantity of tages used and the amount of data being tracked.

  26. efficiency
    January 26, 2011

    how far down in granularity you go affects the cost of using RFID tags but not so much the readers as they get reussed no matter what the quantity of RFID tags go by it.

    since it is mostly humanless intervention the labor should not go up but down.

    The only way it could go up is in using and understanding the info flow to optimize performance of the system.

    At some point diminishing returns law starts to apply.

    As an example you could put a RFID tag on the ARM processor in an Aplle iphone track it before and after it is put on a PCB and then use it to track the PCB assembly to the fianl assembly. then reuse it again for final assembly.

    Then re-used during shipping transit thru to the store.

    then re-used by the consumer as another security feature to authenticate phone and owner to by a coke at the vending machineor down load some music from a kiosk at the airport…

  27. Barbara Jorgensen
    January 26, 2011

    I hadn't thought about one component acting as the tag for the entire end-product. That would certainly reduce costs. There are a lot of logistics issues–such as last-minute component swaps–but the concept makes sense. Thanks for pointing that out, Efficiency


  28. efficiency
    January 26, 2011

    your welcome Barbara,

     I had just thought it up as it was being written 😉

    probbly not patentable but useful, cost effective idea for the supply chain.

    Re-use, Re-use, Re-use…

    all the small ideas and improvements by everyone added up can save ouselves from ourselves.

    But we don't get to see them added up in such a way that shows the cummilitive benifits.

    This idea probably could be implemented in RFID readers to re-use RFID tags quite easily as it progress up the supply chain.



  29. Hardcore
    January 26, 2011

    The main issue in using RFID tags for security , is that it is *not* secure ,So any product that relied on RFID as the basis for secure payment would be fairly well 'Stuffed' before it even left the factory.

    As a result if you designed it into the 'silicon' of the product it would be a 'waste', but there are still patents on the idea.

    Unfortunately the RFID tags are far too easy to clone. Since there is not a central agency issuing security keys, also for correct validation there would be a need for an extended two way secured computational communication channel.

    Currently i suspect the computational processing power of an RFID tag is limited, and there is the issue that RF is usually always 'open'  to anyone wanting to 'sniff' the conversation.

    I suspect the main reason RFID has not taken off in a big way in manufacturing, is that fact that many products are sub-assemblies, and it just does not make sense yet!!

    Consider that if you decided to implement RFID in a production facility, at what point do you do it? , size or cost of the component?

    In the case of an IPhone, do you: RFID the case+the Battery+the PCB+the LCD, the product box+the shipping lot+ the pallet ?

    Then you get into discussions of subcontractors adding in RFID tags to the source materials,  Tape & reel, plastic shipments, metal parts.

    Even at $0.10US each it is already adding >Dollar to the material costs at various stages.



  30. efficiency
    January 26, 2011

    yes that is why i suggested one high value component that gets re-used in following sub-assemblies.

    The rfid tag used for security was a thought in the moment to add one more layer tot he re-use idea.

    It was not given a lot of thought.

    But i did not intend for it to be used by itself but as a 2nd layer of authentication.

    Hardcore you brought up some good points about open to anybody sniffing.

    But this is as easy as it sounds.

    also the power could be controlled to only work a few inches from the point of purchase just as in near field. The exitation source would be very low power as would the rfid reflection. The far field mixing of both would make it very difficult to intercept this signal/

    But I do agree with you that without careful implementation this could be an issue if this was the only authentication messure.

    it at a minimum could be used to cross-correlate the token, credit payment and cell phone user together.

  31. Hardcore
    January 27, 2011

    Hi efficiency,

    Many of the 'secure' RFID systems were specifically designed to be short range, but with the use of special antennas, that has been 'cracked'.

    Also there are 'addon' fronts, basically plastic faces that fit over the 'real'  scanner and feed the signals into slave devices for analysis.

    The bottom line is that RFID as it stands , just is not secure enough to be trusted handling a significant volume of financial transactions.

    If you dig about many of the  'secure' token systems have already been cracked, but they have been/are being prevented from publishing results via a number of high profile  'gag' orders.

    Not that it is going to make the products any more secure since the cracks are well publicized, but it does ensure that less well informed customers keep buying the equipment.

    It is yet another exanmple of' Security via abscurity' having fallen apart, where a manufacturer or group consider themselves so smart as to be able to prevent the combined mentality of several thousand people, just be blocking publication of some data.

    Unfortunately as has been shown so many times in the past , it just does not work, and still they do it.




  32. hwong
    January 30, 2011

    Sooner or later the U.S. will catch up the RFID technology just like in Asia where RFID has been used widely in various places. For example, in Japan RFID chip is installed in the cell phone so that people can use their cell phone to purchase food or beverage in their vending machines. Also in Hong Kong, people  started using Octopus card 20 years ago to pay for all means of public transporation such as the train, MTR, bus etc.   U.S. companies have only considered RFID in the supply chain to keep track of inventory in all stages. But ROI is yet the big question.  What I envision in the future is that we can use our cell phone to pay for purchases just like a credit card. Also we'll use the cell phone to transmit other secured personal data  so that would eliminate having to carry many credit cards and IDs.  OF course we need to think about how to protect privacy and security.

  33. Marc Herman
    January 31, 2011

    I thought it was an Oyster card for London transit. Octopus card is way cooler, though. If it doesn't exist, someone should make it.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.