Using AI in the supply process

Voli Trade doo
Supermarket chain
Using AI in the supply process


Apache Spark
Postgre SQL



AI - Machine learning and artificial intelligence

Software solution based on artificial intelligence and machine learning that facilitates company management



The project aims to optimize supply for both logistics and distribution centers as well as for retail units, enabling them to more easily determine the required quantities for supply. With the introduction of asw:numerus, a revolutionary solution for automating the process of ordering goods, we got a mechanism for automatically generating the necessary replenishments and optimizing the supply process based on data analysis using machine learning. This is extremely important for clients, as it allows them to focus on other aspects of their business, while asw:numerus takes care of automating the process of ordering goods.​​


and The system uses available historical sales and inventory data, as well as basic data: items and their attributes, objects and their attributes, object assortments, supply calendar, additional inventory information. The data is first loaded from the ERP and prepared in the appropriate format so that it can be used for the purpose of training the machine learning model. After that, the model itself is trained, which predicts the demand for each item in each of the facilities, using the defined assortment. Taking into account the latest information on the stock level, the necessary replenishment is calculated. That replenishment is rounded up to the bundle order quantities for each of the items. If there are additional requirements such as marketing stocks or minimum stocks, these are also taken into account when calculating the required replenishment. As information about the supply calendar is available, the system itself initiates all the necessary processes to obtain refills. Calculated required refills for objects and items are available to the user every morning. The user receives these results in the form of a table in the created application, as well as in graphic form through the Tableau tool. Also, users have the possibility to manually check and change the created top-up suggestions. Although the availability of supply calendar information is important, in cases where it is not available, it is possible to manually generate the requirement by selecting the appropriate supply method, supplier, and items. The same mechanism is applied in other situations where there is a need to create a requirement manually. A full integration with the ERP system has been established so that the generated additions in the application are also visible after acceptance. In this way, it is also possible to track which items have already arrived. In addition, asw:numerus informs the user not to reorder the same items that are still on the way.

Related Cases

Integrated information system of public utility company
Technical Sector, Design and Development Sector, Financial Legal Department, Service (Sector) for monitoring technical losses on the network

Integrated information system of public utility company

find out more
Bakery chain in Serbia

Bakery chain in Serbia

find out more