This weekend while I was at the supermarket with Kat, we began analyzing the inefficiency of human-operated checkouts.
At Kat's local supermarket, we typically spend anything from 10 to 20 minutes queuing to get to the checkout. That's a ridiculous amount of time wasted, especially when all you're buying is a litre of milk! The problem consists of a number of factors:
Let's look at this factors individually.
Number of items: The presumed assumption about the number of items is that the greater the number of items, the longer it will take to get through the checkout. This is true above a certain quantity of items. Below this critical mass, the weight of the other factors takes precedence, rendering buys of less than, say, 10 items not affected by the quantity of items.
Cart Unload Time: Some people are hopeless when it comes to unload their carts (in the UK, we call them 'trolleys'). If a single individual is performing the purchase the unload speed is usually quite acceptable. However, there are factors which complicate cart unloading, such as nagging children announcing to the world that they need the bathroom or they must have that Mars bar or they will cry. These complications only seek to distract the unloader from performing their duty. For Kat and I, we have this process down to an art form - we both simultaneously unload the cart which often results in some startling unload times.
Item scanning: How many times have you been at the checkout and the cashier just cannot get the scanner to read the barcode on particular product, so he/she tries, in vain, over and over and over again, until they finally give in? This happens particularly frequently with frozen goods, where a thin veneer of ice reflects and refracts the scanner's laser to the point of a failed read occurring. When this happens, the cashier usually resorts to manually entering the barcode's digits into the POS system. Sometimes, the barcodes, for whatever reason, are just totally illegible and the supervisor trump card has to be called into play. If the supervisor has to go look up the value of the product, this can add in the region of 2 minutes to checkout time. From my observations, tinned goods are the easiest to scan because the regular shape and size of the products.
Post-purchase packing: This is a big of bug-bear for me. Some supermarkets have introduced what they call 'Scan & Pack'. This is where the cashier will scan and then pack the items themselves. In my scenario, this is the most inefficient option to take -- Kat and I are both perfectly capable of packing our own items. Plus there's the factor of there being two of us compared to the one cashier. Two dedicated packers are going to out-pack a ‘scan & pack'-er any day. Quite often, the cashier will ask if they should pack the items. In this scenario, we refuse 100% of the time.
Payment processing: It goes without saying that cash is by far the fastest method of payment. Although, for more and more people (myself included), payment by credit or debit card is far more convenient. Since the introduction of 'Chip & Pin' cards in the UK, electronic payment has increased its efficiency somewhat by removing the need for individuals to remember how to hold a pen, let alone how to use one. Chip & Pin is not without its problems though -- a forgotten PIN is often expedited by the purchaser forgetting which family member's birth date they used for the 4-digit security number.
I hope you can see by now that the supermarket is actually quite a complex user scenario with a lot of mitigating factors whose sole purpose is to increase the stress and frustration levels of those in the queue behind the purchaser. So how can RFID help to remedy this?
By introducing RFID, it is possible to remove two of the above factors out of the equation -- these are: cart unloading and item scanning. RFID does not require an individual to pass the product in front of a laser beam in order for it to register with the POS system. Instead, low-powered radio frequencies are used to provide all the information a barcode scan & database lookup would provide.
By removing the scanning factor, together with the increased adoption of electronic payments, it negates the necessity for the cashier to be present at the checkout. However, this will have a significantly negative impact on the local socio-economics by potentially increasing the unemployment levels should the ex-cashiers be unable to find a replacement occupation.
By adding a card reader and RFID scanner to the cart, a further two factors can be removed formed the equation (namely, payment processing and number of items). With the on-board card reader, preliminary payment processing can begin as soon as the card is inserts. As the shopper moves around the store and items are added or removed from the cart, the RFID scanner tracks these and adds or removes them from the purchaser's bill.
So, as I hope you've seen, the whole supermarket experience can be improved by using RFID. One obvious benefit is to the health of the shoppers – less queuing will result in less frustration and stress. However, this effect may be negated by worrying about the privacy issues raised by tracking the products. Personally, I believe this to be no more worrying than store loyalty cards which already record the entire detail of an individual's purchases. This data subsequently gets analysed with a whole bunch of other data to form a "shopper profile" where stores can fairly accurately predict which products the shopper is likely to buy (and when).