The solution enabled greater accuracy in projections, optimizing decision-making and boosting operational efficiency.
Faced with the need to improve sales predictability and optimize strategic planning, Eucatex struggled to produce accurate demand projections. The traditional process showed a high average error of 19.5%, directly impacting resource allocation and the efficiency of the commercial team.
To address these challenges, MadeinWeb developed a forecasting solution based on machine learning, leveraging historical sales data for advanced modeling and high-accuracy demand projections.
The solution generates monthly forecasts with a 2-month horizon for a portfolio of 1,843 items.
The ingestion and preparation of historical sales data was automated with AWS Glue, with governed datasets stored in Amazon S3 and analytical queries via Amazon Athena.
The ML pipeline is orchestrated by Amazon SageMaker Pipelines, with features managed in Amazon SageMaker Feature Store to ensure consistency between training and inference.
The final production model is an XGBoost, selected from 90 tested configurations — including linear, autoregressive and neural network models.
Training runs via Amazon SageMaker Training, with versioning in the SageMaker Model Registry and recurring batch inference via SageMaker Batch Transform, scheduled by Amazon EventBridge.
Predictive quality and drift monitoring uses Amazon SageMaker Model Monitor with alerts via Amazon CloudWatch.
The main steps of the project included:
Data mapping and ingestion: collection and automation of relevant information sources.
Predictive model development: 100+ variables created and 90 models tested and validated.
Interactive dashboards: creation of panels to visualize metrics and monitor performance.
Continuous optimization: adjustments and improvements to maximize forecasting accuracy and efficiency.
By implementing the demand forecasting solution with MadeinWeb, we are beginning a significant transformation in our inventory management. With this partnership, we hope to forecast demand more accurately, ensuring our products are always available at the right time and in the ideal quantity.
— Eliezer Ferraz, Business Intelligence Supervisor – EucatexWith this new approach, we are confident we can optimize our processes, better meet market demands, and make our operations increasingly efficient and agile. MadeinWeb's expertise gives us the assurance of a positive and sustainable impact on our business.
— Marcio Roberto Crespo Candido, Corporate IT Manager – Eucatex
Amazon Web Services (AWS) is the world's most broadly adopted and comprehensive cloud platform, offering over 200 fully featured services from data centers globally. Millions of customers, including the fastest-growing startups, largest enterprises and leading government agencies, are using AWS to lower costs, become more agile and innovate faster.
With AI agents applied to the development cycle, DM saved 337 hours, boosted Product and Quality productivity by over 40% and standardized deliveries in just a few months.
ONR created IARI, a generative AI platform that automates the extraction and standardization of property registration data across real estate notary offices, with plans to reach more than 3,000 registries.
Tenda automates contract generation and eliminates 90% of the manual process
We use cookies to improve your experience on our site. Essential cookies are necessary for the site to function. You can manage your preferences or accept all cookies. Learn more
