A Novel Application of Tensor Networks for the Investigation of Design Optimization Tools in the Marine Domain

Abstract

Traditional optimization methods often struggle to map the unique interactions between design variables, operational constraints, and performance objectives. Tensor networks, a mathematical framework rooted in quantum physics, address this challenge by providing a tool to model state relationships within multidimensional data structures. In the context of bulk carrier synthesis and optimization, tensor networks enable the simultaneous analysis of multiple constraints and their interactions via a state space representation. A state space representation offers a holistic understanding of the optimization landscapes by providing insights that add to traditional optimization analysis techniques. This paper presents a methodology for converting the optimization problem into multiple tensor network representations, details the implementation of tensor
network algorithms, and showcases implementation results. The findings underscore the capacity of tensor
networks to provide a deep, data-driven understanding of complex optimization landscapes, thus enabling
novel decision-making opportunities.