Intelligent Systems for Industry, Supply Chain and Environment
A.Y. 2021/2022
Learning objectives
The aim of this course is to introduce students to methodologies and techniques for intelligent systems for monitoring and control of industrial, environmental and supply chain applications, typically based on artificial intelligence techniques.
Expected learning outcomes
At the end of the course, the student will have acquired the ability to design, train, measure the performances and compare intelligent systems in order to achieve functionalities such as monitoring and control in industrial, environmental and supply chain applications.
Lesson period: Second semester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course cannot be attended as a single course. Please check our list of single courses to find the ones available for enrolment.
Course syllabus and organization
Single session
Responsible
Lesson period
Second semester
More specific information on the delivery modes of training activities for academic year 2021/2022 will be provided over the coming months, based on the evolution of the public health situation.
Course syllabus
The course presents artificial intelligence techniques for the design, training and the effective deployment of intelligent systems in a wide range of applications.
After the description of the general design techniques for intelligent systems, ranging form the creation and management of the training dataset to the choice of the best models and training methods, the course presents different tools and environments for the implementation of the systems.
The course presents various examples with code to be tested on various environments.
Main topics covered:
· Artificial intelligence, intelligent systems and their application.
· Deep learning systems: design, training, validation and optimization.
· Advanced techniques and best practices. Data augmentation and transfer learning.
· Main environments for the creation of intelligent systems and deep learning (with specific examples using TensorFlow and Keras).
· Application of deep learning techniques for applications based on images.
· Application of clustering techniques and classification for the creation of intelligent systems.
· Automatic feature extraction from unstructured data, structured data and images.
· Information fusion.
· Intelligent systems for prediction.
· Application to ambient Intelligence.
· Application to signal and images: feature extraction, fusion of multi-sensor data. Study of industrial cases.
· Application to intelligent sensors and sensor networks. Acquisition and processing of sensor measurements. Study of environmental and ambient intelligence examples.
· Prediction, monitoring and control for applications for industry and the environment and the supply chain. Quality prediction.
· Applications of intelligent system for industrial processes, industrial automation, robotic systems, complex products, power distribution grids, automotive and transport systems and the supply chain.
A detailed list of topics of each lesson is presented and regularly updated on the course site.
After the description of the general design techniques for intelligent systems, ranging form the creation and management of the training dataset to the choice of the best models and training methods, the course presents different tools and environments for the implementation of the systems.
The course presents various examples with code to be tested on various environments.
Main topics covered:
· Artificial intelligence, intelligent systems and their application.
· Deep learning systems: design, training, validation and optimization.
· Advanced techniques and best practices. Data augmentation and transfer learning.
· Main environments for the creation of intelligent systems and deep learning (with specific examples using TensorFlow and Keras).
· Application of deep learning techniques for applications based on images.
· Application of clustering techniques and classification for the creation of intelligent systems.
· Automatic feature extraction from unstructured data, structured data and images.
· Information fusion.
· Intelligent systems for prediction.
· Application to ambient Intelligence.
· Application to signal and images: feature extraction, fusion of multi-sensor data. Study of industrial cases.
· Application to intelligent sensors and sensor networks. Acquisition and processing of sensor measurements. Study of environmental and ambient intelligence examples.
· Prediction, monitoring and control for applications for industry and the environment and the supply chain. Quality prediction.
· Applications of intelligent system for industrial processes, industrial automation, robotic systems, complex products, power distribution grids, automotive and transport systems and the supply chain.
A detailed list of topics of each lesson is presented and regularly updated on the course site.
Prerequisites for admission
None.
Teaching methods
Frontal lessons
Teaching Resources
Course site: http://fscottiisie.ariel.ctu.unimi.it/v5/home/Default.aspx
Slides of the single lesson published on the course site
Handouts published on the course site
Slides of the single lesson published on the course site
Handouts published on the course site
Assessment methods and Criteria
The exam is composed by a single compulsory closed book test to be filled using the PCs in the laboratory requiring the solution of exercises and questions about the application and the design of intelligent systems, and multiple choice questions to verify the knowledge of theoretical notions of the same type and difficulty of the cases proposed and discussed during the curse. The duration of the exam is about 1 hour. The grade is expressed in thirtieths.
Professor(s)
Reception:
By appointment (via email)
Computer Science Department, Via Celoria 18 - 20133 Milano (MI), Italy