h a l f b a k e r yReformatted to fit your screen.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Existing "smart shelves" for retail inventory at worst can only tell the weight upon them and at best can only interface with a traditional inventory management system. That is to say they report to the story which items are on them and the store's centralized inventory system does the rest. This is
kludgy, inefficient, slow, and inflexible. For shelf space to be reallocated a human has to make decisions about what to put where, a human has to manually enter the new data, a human has to reallocate space, and a human has to stock the shelf or program a robot to do so. Price changes are centralized and change slowly, in response to -- in some cases -- years old data. It's just so last century, with a thin veneer of modernization.
This smart shelf knows what's on it, but it doesn't report inventory numbers for stocking. Instead it keeps track of how fast things are selling using a continually updated branching tree algorithm. If its predictions are wrong long enough it prunes back the beliefs that led to those predictions and generates new ones. Degree of change and time of change are also modified by a similar genetic algorithm. It does not need an enormous complexity, it could run on a couple of Pis.
The shelf is given money by the store. The shelf purchases stock from a central inventory system which does not need to know what's happening on the shelves, at least not for day-to-day operations. The shelf knows its profit margin and can bid for higher or lower margin items. The shelf knows how fast it empties and bids for restocking labor to be performed by robots or humans. All shelves in the store compete to be as profitable as possible. In the case of inventory shrinkage or negative margins a shelf can go broke, in which case an exception is created for human intervention. Otherwise these shelves function as mini-stores in their own right.
So the shelves containing similar items are likely to still group together since that's the best way to get things purchased. The shelves set their own prices, inherently maximizing margins. The shelves will collectively exhibit the emergent behavior of changing item locations until the store is set up in the most profitable possible configuration without a single human intervention necessary.
The net effect of this is the equivalent of changing from a centralized economy to a decentralized one, all the way down to the shelf level. Next up: the laundry soap that sells itself
Bodegas
https://www.theguar...odega-corner-stores - Fill a warehouse with these - tada [mylodon, Sep 27 2022]
[link]
|
|
Anytime there was a supply shortage or change of season the entire layout would change, and I'd never find the eggs. i have a hard enough time finding eggs as it is. |
|
|
I'd have to send my artificial artificial intelligence bot running off a set of complicated mad libs to do the work for me. |
|
|
However, maybe that would solve itself out. Even the local markets have the food trucks always stay in the same place. Nobody seems to be trying to fight to get close to the door. I don't know if this is by design by the market organiser, or if it is beneficial to know where to get your churros. |
|
|
But it's an interesting idea to take market-based economics to the shelf level, and to see where the winners and losers would be. Would the invisible hand of the market keep the eggs in the same place? |
|
|
If adjacent shelves competed with each other, offering slightly different prices on the same items, this might benefit the customer but would work against the interests of the shopkeeper. |
|
|
You've heard of Elf on a Shelf... well now there's Shelf that's an Elf. |
|
|
"Master has given Dobby another package of socks." |
|
|
you read that in his voice, admit it |
|
|
"Dobby... is... on sale at 50% off!" |
|
| |