The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B4-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B4-2022, 419–425, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B4-2022-419-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B4-2022, 419–425, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B4-2022-419-2022
 
02 Jun 2022
02 Jun 2022

EXPLORING QUANTUM COMPUTING POTENTIALS IN SOLVING A COMBINATORIAL OPTIMIZATION PROBLEM TO MINIMIZE EXPOSURE TO COVID-19 DURING A CITY JOURNEY

A. Nourbakhsh1, M. Jadidi1, and K. Shahriari2 A. Nourbakhsh et al.
  • 1Geomatics Engineering, Lassonde School of Engineering, York University, 4700 Keele Street, Toronto, ON, M3J 1P3, Canada
  • 2Intelligent Data Analytics, Toronto, ON, M9A 1R9, Canada

Keywords: Vehicle routing problem, Quantum computing, Quadratic unconstrained binary optimization, Quantum annealing

Abstract. COVID-19 is an airborne virus that can be spread directly or indirectly from one person to another. Spreading the virus strongly depends on the location and time and hence, a Spatio-temporal event. Moreover, traffic congestion will increase the spread of the virus not only because of the vicinity but also because of increased temperature and humidity in these spaces for a short or long time. This paper introduces a vehicle routing optimization model to reduce COVID-19 exposure risk during a city journey by solving it as a quadratic unconstrained binary optimization problem on a quantum annealing computer. Indeed, the objective of the COVID-19 prevention optimization problem is to minimize the risk of exposure for a given set of road users between origins and destinations. Microsoft Taxi data from the city of Beijing have been used to simulate road users’ movement. The problem has been run onto three different solvers. One of the solvers is executed on classical computers, and two other solvers are executed on hybrid quantum solvers. Hybrid solvers return the solution within less than 0.03 seconds on quantum processing unit time. However, the results will be returned at least 5 seconds after the execution in the classical solver. It is worth mentioning that, as there is no direct access to the quantum computers, it is hard to compare the results on the same scale as the queries will go on a queue in D-wave quantum computers. Applying the proposed model on the trajectory data shows a better distribution of the vehicles on the road network.