ai-resources
AI-based Demand Forecasting
AI-based Demand Forecasting supports production managers in the phase of acquisition of raw materials. It includes an on-demand forecasting module, which is based on Machine Learning techniques including Extreme Gradient Boosting Regressor.
CUSUM RLS filter
CUSUM RLS filter contains a change detection algorithm for multiple sensors, using the Recursive Least Squares (RLS) and Cumulative Sum (CUSUM) methods. This aims to detect abrupt changes on the measurements recorded by a set of sensors.
AI4Manufacturing Toolkit
A collection of operational technologies, data analytics tools and platforms, designed to provide support to system integrators and technology adopters to create new AI-based applications.
Reinforcement Learning for Assembly Line Balancing Enhancement
The Reinforcement Learning for Assembly Line Balancing Enhancement is an application of a Neural Network trained with Reinforcement Learning method aimed to dynamically assign production resources to tasks in a production schedule during its execution.
Supervised real-time 2D-based Object Detection system
A flexible tool able to identify and localize in real-time the best object to pick in scene with a multitude of overlapped identical objects.
Process monitoring and optimization toolkit (POCAS)
Set of machine learning and AI routines for advanced process monitoring, analysis and optimization
EASY OPT
EASY OPT is a friendly tool able to prescript optimal setpoints of process control parameters that will help industrial processes staff to take better decisions for the improvement of their industrial processes.
Digital Twin For Predictive Maintenance
A machine learning data pipeline to perform predictive maintenance in an industrial context
AI-based Production Scheduling
The AI-based Production Scheduling supports production managers in the organization of production. It uses Metaheuristics techniques to work orders sequencer based on Metaheuristics.
D2 Anomaly Analyzer and Predictor
Framework for discovering, analysing and predicting anomalies in high dimensional process spaces. It is main advantages are related to supporting multivariate process control and resolving unknown situations.
Intelligent Computer Vision for Digital Twin
The Intelligent Computer Vision for Digital Twin allows for Cyber-Physical Systems, particularly robotic systems, to rely on their perception systems to keep a digital representation of a manufacturing environment updated through the usage of Cloud, Edge and Local AI capabilities