Frontiers in Physics (Mar 2023)
Quantum annealing for the adjuster routing problem
Abstract
In the event of a disaster such as an earthquake, insurance companies basically conduct on-site witnessing. Depending on the scale of the disaster, hundreds of adjusters are dispatched from each office to the affected buildings per day. In such cases, which adjusters will witness which buildings and in what order must be determined, and the route must be optimized to conduct efficient witnessing. In this study, we define this witnessing route decision as an optimization problem and propose the adjuster routing problem (ARP). The ARP can be viewed as an extension of the vehicle routing problem (VRP). We introduce constraints not to be considered in the usual VRP, such as adjuster-building matching and satisfying the desired time. The VRP is an NP-hard optimization problem and is considered difficult to solve on a classical computer. Therefore, we formulated various constraints in QUBO so that quantum annealing can be applied to the ARP. In addition, we conducted numerical experiments with D-Wave. The ARP is a real problem, and our research provides a new example of applications of quantum annealing to real-world problems.
Keywords
