Scalable Machine learning for Design Support System

Degree Course: 
Electronic/Computer science engineer (Information-engineering track)
Department of Information Engineering - UNIVPM

iDelph aims to be the first Design Support System based on Machine Learning model. It can be seen as a software based on artificial intelligence able to learn, correlate and interpret all the parameters of a database, representing the characteristics of a product or production process, in order to predict new possible versions. iDelph aims to use and maintain over time the know-how of designers supporting the design phase by instantly proposing possible novel solutions while reducing effort in terms of time and cost. iDelph is able to analyze not only numerical parameters (discrete/continuous), but also heterogeneous non-numerical data, such as colors, shapes, geographical or market data. Differently, from the traditional simulation software that employs a model-based approach, the iDelph predictive
system is based on a data-driven approach. Every time you add a new result or a new variant, the system is able to learn and improve, increasing its predictive potential over time. The iDelph project was created to provide innovative support that allows the optimization of design in manufacturing, engineering and technical, but is also applicable to other sectors such as data analytics, marketing, etc.
In this context, the candidate will be required to investigate and develop scalable machine learning model. This model, after learning from a known dataset, is able to predict some parameters without the a priori knowledge of input and output.

Aim: Development of scalable Machine Learning Model for support the design of new products in the Industry 4.0 scenario

Supervisors: Prof. Emanuele Frontoni, Luca Romeo, Gianluca Bocchini (Digital Product Specialist with Xelexia s.r.l)

Start: Sept, 2018

Expected graduation: Feb/July, 2019

Skills that will be acquired: Programming (MATLAB)

Academic Tutor: 
Emanuele Frontoni