Introduction and models for the Travelling Salesman Problem (TSP). Code setup and input parsing in C (TSPlib format). Basic ILP model and its solution through branch-and-cut (using IBM ILOG Cplex software). Compact models and their implementation in Cplex. Benders-like solution scheme using Subtour Elimination Constraints (SECs). SEC separation using Cplex's callbacks. Computational comparison of alternative exact approaches to TSP through performance profile plots. Heuristics and meta-heuristics for TSP: constructive heuristics, k-opt refining, tabu search, variable neighborhood search, simulated annealing, genetic algorithms etc. Writing a scientific paper (in Latex) on the implemented methods and on the corresponding computational performance. Teaching scheme: Front lessons (plus homeworks) via ZOOM at Fischett's classroom https://unipd.zoom.us/j/364970024 Exam: oral discussion of a thesis developed during the course (2-person groups). Bibliography: Notes provided by the teacher and available at http://www.dei.unipd.it/~fisch/ricop/RO2/