SELF SORT -- Automated Stocking and Sorting System for Inventory and Retail Control

How does it work?

-Controls stocking

-Adheres to planogram

-Tracks total inventory through inflow AND outflow from coolers

-Assures correct product placement

-Eliminates human error while maintaining customer experience

Employees (or customers) insert bottles at the top left hand tube. The bottles then slide to the back and a 360 degree barcode scanner reads what type of product is inserted. Depending on the product, a certain configuration of trapdoors open and close, allows for the product to be placed on the correct shelf. It slides forward on the shelf as far forward as possible. A turnstile system allows one drink to slide through to the front of the cooler, while preventing any product from being pushed back into the cooler (this ensures that product inflow and placement can be controlled by our input system and/or employees).

What does it measure?

Our system measures the inflow and outflow of inventory into the coolers by using a barcode scanner at the input and a controlled planogram with row counters at the output. This information can be exported to the company and used for better stocking and inventory control, as well as for sales optimization and store-by-store customization.

Does the retailer/bottler/distributor have to do anything different to make your solution work? (if yes, please explain)

For the primary step with the gate and button, there are no changes required on the cooler. The solution is modular, flexible to any number of rows, and fits all cooler dimensions. These can easily be installed by the retailer. For the complete solution, the bottler and distributor do not need to make any changes. The retailer will need to stock the machine using the mechanism provided, and new shelf racks will be needed in order to install the system.

What materials does your solution require?


-Raspberry Pi/Other Internet connected circuit


-Plastic moulded scaffolding

Best guess on cost to implement?

<=$50 per cooler, not including industry pricing and economies of scale. This estimate was based off of commercial pricing for the products we used. On an industry wide scale, Self Sort can be manufactured for much much less because our cost came from commercial grade mechanics and supplies.

Challenges you/your team ran into?

Strength of servos and alignment of trap doors. Duct tape does NOT fix everything. Getting a hardware prototype finished in less than 48 hours is far more challenging than it sounded when we first started.

What you/your team learned?

We learned that nothing ever works the first time you try (and to take more time to do these questions - haha!) We stepped through several different iterations of a self stocking design, and learned that strategic mechanical design can overcome difficulties in programming. Using gravity as the movement mechanism allows the system to use less electronics and mechanics. We also focused on the mechanics more than the software, but we did our best to leave plenty of room for the potential growth in that area.

As a group of Georgia Tech Engineers who have participated in research and senior design, iterating through the complete design process in 48 hours was an awesome challenge. It allowed us to apply the physical skills we have learned in class to a real life challenge.

Next steps?

Moving forward, we would like to produce a better prototype with more stringent specifications so we can simulate real world use of our solution. We would also like to flesh out the software side of things so that we can really take advantage of the big data gathered. There is a lot of potential to optimize planograms on a per-store basis -- all this development could come with more focus on the software.

Anything else you want to add?

Self Sort is novel because it circumnavigates stocking problems and better serves the planogram. It also has been designed to be incorporated into existing machines in two steps: an immediate and effective solution by controlling and measuring outflow that could be implemented immediately, and a larger all-encompassing solution that controls inflow as well. Also, many design choices have been made considering customer experience -- the control of inflow and outflow prevent customer randomization and still allow for the same buyer experience. Furthermore, corporate control over retail success is a huge plus. The benefits of this design are endless as the concept continues to be developed, and our solution solves beyond the problem of inventory count and management. We really enjoyed working on this project and would love to see it progress further.

Built With

Share this project: