The client


Topex Group is a prominent manufacturer of tools and equipment solutions for the construction industry, distributing their products by the largest retailers like Castorama, Leroy Merlin and OBI. Operating in multiple markets, including Central and Eastern Europe, Topex Group offers a wide range of high-quality products and services.

The Topex Group Tools and Equipment division supplies a comprehensive selection of tools, machinery, and accessories, catering to the productivity and efficiency requirements of construction sites.


Additionally Topex Group provides building materials sourced from reputable manufacturers. Their extensive range includes structural components and finishing products, as well as training and education services to support professional development in the construction sector.




The problem


Every month the forecast department of the Topex group on the basis of its expert knowledge, market analysis, trends, weather, number of construction sites in the country, had to manually try to forecast the necessary stock level for each of its products to meet the demand in the following months. The process required maintenance of the team of highly skilled experts resulting in high operating costs.


Additionally, because of the scope of the tasks with thousands of products and variety of factors impacting forecasts the risk of error was very high, especially in the long term. This translated into supply chain disruptions, causing a loss of revenue and perceived reliability for the Topex group.



The challenge

  • The amount of products for which demand had to be forecasted. Our system had to compile recommendations for stocking of over 6 000 products.


  • Multitude of constantly changing factors and parameters such as historical sales trends, seasonality, deviations e.g.: periodic increase in orders, that had to be taken into consideration during forecasting.


  • The need for proper model that could handle required parameters and relatively low cast as well as scale and adapt to potential changes and impactful events that could shape the market demand.



The solution


After thorough testing we chose an analytical specialized framework, available under an open-source license. We took advantage of the configuration options to adapt the most optimal mathematical model to the specifics of Topex group sales.

The solution is integrated with the data through ADF pipelines and the ERP system provides AI based forecasts into it. Forecasts are generated for the next 10 months while direct recommendations for the next month.


We used the cloud as a base for our solution to ensure its scalability. The great advantage of the cloud is the ability to run a large number of servers just for the duration of the task – providing great flexibility.


Thanks to this approach, we have reduced the forecast calculation time to 2-3 h instead of several days.





Databricks: For application deployment


Azure Data Factory: For data processing and preparing pipelines through which the prediction generation process is triggered.


Azure Key Vault: For Application security verification


SQL Server: For On-prem database


Azure Blob Storage: Data in-cloud storage.


Power BI: For presentation layer and as self-service, allowing the business user to view multiple interactive dashboards, gaining concise, actionable insights.


Forecasting department

The forecasting process was fully automated and supported by innovative technology, resulting in elimination of human error and reduction of monthly execution time to just 1 hour. This not only reduced repetitive and labor intensive task of providing recommendations to all 6, 500 products but also allowed the department to focus on special and uncommon cases, increasing its efficiency.


Company image

Thanks to our solution forecasting errors were reduced by 50%, this resulted in 30% less unfulfilled orders – not only increasing revenue, but also improving reliability of Topex Group as supplier. This in turn caused its image as a business partner and as a brand to improve, causing an upturn in the amount of future contracts.


The bottom line

Due to improved forecasting more orders were fulfilled and less superfluous products were ordered. This meant higher revenue and in conjunction with a reduced amount of forecasting experts needed, led to much lower costs.