PEPITe develops DATAmaestro, a no-code platform that utilizes machine learning and advanced analytics to enhance industrial operations by optimizing energy efficiency, productivity, and product quality. The software enables manufacturers to effectively collect, analyze, and visualize data, addressing issues of waste and resource underutilization in manufacturing processes.
Funding
Funding not disclosed
Founders
Product
Problem
Many industrial operations struggle with inefficiencies related to energy consumption, productivity bottlenecks, and inconsistent product quality due to underutilization of available data and lack of real-time insights. Diagnosing root causes of variability and predicting performance drifts in complex manufacturing processes can be challenging without advanced analytical tools.
Solution
PEPITe's DATAmaestro is a self-service advanced analytics platform designed for engineers to optimize industrial plant performance. The platform connects directly to various data sources, including historians, LIMS, MES, ERP, IoT sensors, and PLCs, to centralize data collection and storage. DATAmaestro provides tools for data preparation, visualization, and machine learning model building, enabling users to diagnose performance issues, predict outcomes, and optimize processes. Real-time dashboards allow for continuous monitoring and decision support, facilitating timely adjustments to maintain optimal performance and reduce environmental impact. The OPTImaestro methodology combines the DATAmaestro platform with the knowledge of plant personnel to maximize the impact of advanced analytics on the production process.
Target Audience
The primary users are engineers, operators, data scientists, managers, and R&D personnel in both small and large manufacturing enterprises.
Features
- Direct data connection to historians (IP21, PI, Wonderware, eDNA), LIMS, MES, ERP, RDBMS, IoT sensors, PLC, and DCS
- Centralized data lake for high-speed ETL and data preparation
- Comprehensive environment for machine learning with algorithms tailored for industry, including decision trees, random forests, and neural networks
- Real-time dashboarding widgets for monitoring process performance and asset health
- Web API for integration with existing applications
- Flexible deployment options, including cloud and on-premise
- Thermophysical properties library for feature engineering