During the pandemic period, the e-commerce sector experienced unprecedented growth, with a significant increase in demand from consumers who preferred to shop online to avoid the risk of contagion.
This was also the case for one of our long-standing partners, a fashion e-commerce that saw a sharp increase in orders during the pandemic:
in 2020 alone, sales increased by over 32%. The rapid increase in demand initiated a virtuous cycle of growth but also brought to light new logistical problems.

Order processing and shipment preparation procedures, in particular, were becoming increasingly difficult to manage: originally handled directly by warehouse operators, they involved manually collecting orders from marketplaces and processing each request one by one. This meant that operators were responsible for monitoring marketplaces, responding to users, updating inventory, and creating and sending documents related to order fulfillment.

In 2020, however, with the rapid increase in demand, operators could no longer keep up with the workload: the e-commerce began to experience serious slowdowns in workflow, and errors and mishaps became more frequent. Delivery delays increased, and overall operational efficiency significantly decreased.

For this reason, our partner decided to turn to us to find a solution that would automate and simplify logistics workflows as much as possible.

Our team developed software for automating workflows and logistics processes, based on cloud microservices and integrated with Machine Learning algorithms, to fully digitize and automate sales logistics management.

On one hand, the software manages the collection and processing of order data, automating the process of entering and managing request information and robotizing inventory management. On the other hand, it uses advanced Machine Learning algorithms to perform 'smart assignment' of orders to the most suitable couriers and then proceeds with the automatic execution of shipping processes.

First, the platform automatically receives and collects 'raw' orders from the various marketplaces on which the company operates: the proprietary e-commerce and other platforms such as Amazon and Meta.

It then automatically processes the order: the original identification code is converted, and additional identification codes for the warehouse and couriers are added to the initial data. The product information is enriched with other useful metadata (regarding the actual weight of the product, fragility, other particularities) and then saved in the database.

Once the database is populated, the courier selection for shipping the goods is done automatically. A Machine Learning algorithm collects and evaluates the metadata related to the order and calculates, based on certain criteria (such as weight, product fragility, shipping location, and history of previous shipments), which courier is most suitable for the specific case.

The designated company operator can also choose to initiate the shipping process at any time: the platform is integrated with various shipping services (TNT, BRT, ...) through specially developed APIs, and the shipping request can be automatically transmitted to the courier selected by the algorithm. This step allows the shipping procedure to be immediately initiated on the courier's side, while on the company's side, it ensures that the necessary documentation is already completed and ready for shipping.

Once the shipping requests are sent, the platform sends a notification to the marketplaces (to confirm the order has been taken over) and one to the warehouse (to indicate the goods to be prepared); meanwhile, the inventory and warehouse status are automatically updated.

Finally, there is the issue of packaging and labeling the goods to be ready for shipment. The goods are processed at multiple stations, creating potential overlap between operators: orders processed at different stations could involve the same product.
For this reason, we designed and developed a rather complex parallelization solution based on AI - which manages to avoid potential overlaps and ensure a smooth packaging process.

Automating the various operations of order intake, processing, and management has drastically sped up operations in company logistics, increasing the productivity of individual operators and enhancing the overall request handling capacity.
Thus, customer satisfaction has increased, and the company's market competitiveness has significantly improved.