Automatic container identification

Industry

Services

Technology

Storing a company’s merchandise can be a bit challenging, because no matter if you have a large space to store the products, you must have an effective organization system so that no items are lost.

Containers are the most widely used storage instruments by companies, but these must also be organized correctly, to avoid waste of time and effort of staff.

Aquiles Solutions can develop digital tools for companies that allow the correct storage of their goods, so companies can safeguard their products in a correct and orderly manner.

 

Approach to the problem

The successful case of this company, which is dedicated to transport and logistics, was the request for solution to the problem of storage of its empty containers, because they work with different shapes and sizes of them.

To group the same types of containers in one place, operators had to go through the company’s warehouses to place the containers where they belonged, wasting time and money on doing so.

The company’s request to Aquiles Solutions was to develop a tool that would allow automatic recognition of the container that a forklift was loading, so that the operator would receive the location where it hat to move the container to.

The application should be able to:

  • Automate container class identification.
  • Reduce unnecessary forklift travel.
  • Improve control of the number of containers in the area.

 

Aquiles Solutions Proposal

Aquiles Solutions proposed the development of custom software for companies based on  artificial vision, to find the kind of container that the forklift loaded and that would let the operator know where to take it.

This software was developed as follows:

 

Phase 1: Taking information and developing the tool

The forklifts were equipped with a simple board computer and a camera that took the necessary data to develop the classification of the containers.

A tablet was also installed, where the type of container that was being loaded was introduced.

 

Phase 2: Prototype system

A prototype system was developed. This prototipo was installed in just one forklift, to do system tests in real situations.

The tool communicated with the company’s warehouse management system, providing the type of container that was detected and obtaining the location of the container, showing it to the user.

 

Phase 3: Deploy to all forklifts

The application was extended to all forklifts in the warehouse and a functionality was developed that allowed to automatically collect examples of poorly classified containers from the different forklifts, so that they could be retrained and thus improve the model.

DOWNLOAD THE PDF

More cases

Smart Roundabout based on Deep Reinforcement Learning
Purchase Order Reader: Automated Data Extraction
AI to support the introduction of safety observations