Dissertation defence (Applied Mathematics): MSc Alaleh Maskooki
Time
14.4.2023 at 12.00 - 16.00
MSc Alaleh Maskooki defends her dissertation in Applied Mathematics entitled “Optimal routing in dynamic and uncertain networks: an application to maritime logistics” at the University of Turku on 14 April 2023 at 12pm (University of Turku, Main Building, Säästöpankki lecture hall, Turku).
Opponent: Professor Alberto Ceselli (University of Milan, Italy)
Custos: Professor Marko M. Mäkelä (University of Turku)
***
Summary of the Doctoral Dissertation:
The travelling salesman problem (TSP) and its variants have widely been studied in logistics operations and supply chain systems due to the adaptation of the problem definition to such real-world applications. The classical TSP is defined as follows: "Given a list of nodes and the distances between each pair, what is the shortest possible route that visits each node exactly once and returns to the origin". Despite of its simple definition, the problem belongs to the class of computationally hardest among all the optimization problems. The dynamic TSP, which is the base model for the present research, is a TSP with time-varying parameters.
The topic of this research is motivated by a case study on the management of an autonomous or a driven vehicle based on path prediction in a dynamic and uncertain environment. In our case, the dynamic network is defined by uncertain locations of targets to be visited by the vehicle over time. Based on the assumptions and criteria of the case-study, a variant of the problem called moving-target TSP is defined. The models proposed in the dissertation consider the problem from both deterministic and stochastic point of view, and provide mathematical approaches that support decision making for logistics operations. Experiments confirm that high-quality solutions can be found by the proposed methods in a reasonable time. Thus, the methods are highly promising and useful in practical applications.
Opponent: Professor Alberto Ceselli (University of Milan, Italy)
Custos: Professor Marko M. Mäkelä (University of Turku)
***
Summary of the Doctoral Dissertation:
The travelling salesman problem (TSP) and its variants have widely been studied in logistics operations and supply chain systems due to the adaptation of the problem definition to such real-world applications. The classical TSP is defined as follows: "Given a list of nodes and the distances between each pair, what is the shortest possible route that visits each node exactly once and returns to the origin". Despite of its simple definition, the problem belongs to the class of computationally hardest among all the optimization problems. The dynamic TSP, which is the base model for the present research, is a TSP with time-varying parameters.
The topic of this research is motivated by a case study on the management of an autonomous or a driven vehicle based on path prediction in a dynamic and uncertain environment. In our case, the dynamic network is defined by uncertain locations of targets to be visited by the vehicle over time. Based on the assumptions and criteria of the case-study, a variant of the problem called moving-target TSP is defined. The models proposed in the dissertation consider the problem from both deterministic and stochastic point of view, and provide mathematical approaches that support decision making for logistics operations. Experiments confirm that high-quality solutions can be found by the proposed methods in a reasonable time. Thus, the methods are highly promising and useful in practical applications.
University Communications