Predictive Mainteinance
Changed to machine learning for ephylectic seizure prediction candidate: Alberto Pitti Machine learning for fault prediction in mechanical engines
Shortest paths in multi-modal dynamic networks
Development of efficient shortest path algorithms for real life path planning problems involving different means of transportation, time, multiple objectives, big data Candidate: Andrea Nencini Requirements: advanced coding skills, preferably C++
Semi supervised learning: Branch and Bound
Implementation of global optimization algorithms, preferably in python, to trai a SVM in which some of the data has no label Many algorithms exist for S3VM – exact (branch and bound) and heuristic (global optimization). We aim at implementing some
Data driven inventory management
This thesis is on numerical experiments for optimal ordering decisions based on (big) data avaialble – a generalization of the classical newsvendor problem which is data driven and robust with respect to the uncertainty of demand Candidate: Sadhana Rebehi