活动/讲座时间：2024-06-13 10:00

活动/讲座地点：犀浦校区交通运输与物流学院 417学术报告厅

活动/讲座嘉宾：Marcella Samà 副教授

嘉宾介绍：

Marcella Samà is an Associate Professor in Operations Research at Roma Tre University. She received a Bachelor’s Degree in Computer Science Engineering, a Master’s Degree in Automation and Management Engineering and Ph.D. in Operations Research at Roma Tre University with the thesis “Models and algorithms for the real-time railway and air traffic flow management problems”. Her research focuses mainly on designing and developing models and algorithms aimed at solving scheduling and combinatorial problems for real-world application in public services problems, with a specific regard to transit and logistics. She works on decision support systems for real-time management problems in the railway and air traffic flow sectors. More in detail, she specializes in implementing and integrating open and closed-loop deterministic solvers that include advances OR/MS methodologies, their application on train/aircraft real-time rerouting and rescheduling when delays and disruption make initial operation plans unfeasible, route subsets selection and task assignments, as well as the evaluation of risk of additional delays of a given plan when operations duration are uncertain.

主要内容：

In case of unexpected disturbances and disruptions that may affect the normal course of daily operations, the real-time railway traffic management problem is the problem of detecting and solving time overlapping conflicting requests done by multiple train on the same track resource in order to produce a new disposition timetable. This is an NP-hard problem, often studied in the literature with different perspectives and from multiple angles. Given the real-time nature of the problem, different modelling and algorithmic choices have been taken into account to either speed up or simplify the solution process: some approaches limit the size of the problem, intervening on the granularity used to model infrastructures, the time horizon chosen, the number of variables considered; others focus on the solution process to properly drive the search. This talk aims at providing a general overview of the problem and on popular modelling and algorithmic choices used to make the problem easier to solve.