Demonstration of a compatibility-based childcare support service using quantum annealing
- 1Department of Physics, Institute of Science Tokyo
- 2Graduate School of Information Sciences, Tohoku University
Abstract
In contemporary Japan, isolated parenting has become a serious social issue, increasing psychological stress on parents and potentially affecting children’s development. Existing childcare support services tend to focus on physical assistance, while psychological support and community connections remain insufficient. To address this gap, we developed a service that connects parents with senior community members who have parenting experience, aiming to provide psychological support and foster intergenerational exchange. Achieving high-quality matching requires considering pair compatibility, balancing supporter workload, and handling scheduling constraints, which can be formulated as a combinatorial optimization problem. We represented this matching problem as a Quadratic Unconstrained Binary Optimization (QUBO) model and evaluated a quantum-annealing-based sampling pipeline against simulated annealing (SA) and the exact mixed-integer optimization solver Gurobi. In randomized benchmark instances, the QA-based pipeline generated high-quality and diverse matching candidates, particularly for larger instances; at n=15 , the mean best relative error was 1.94% for QA with post-processing and 7.45% for SA. Furthermore, a proof-of-concept field experiment conducted in Sendai City, Japan, demonstrated that the framework can generate multiple high-quality matching candidates, enabling flexible scheduling in real-world operations.
Related
- "It just adds that other dimension to your decision making": findings from a UK study with adult safeguarding practitioners of a pilot forensic service to support adult safeguarding.
- Infection prevention training and adherence to infection control practices in childcare and early childhood education facilities during COVID-19: a nationwide cross-sectional study in Japan
- On-site WBGT at infant height in a childcare facility during peak summer: sun-exposed areas exceed national published values and shaded ground-level environments reach extreme temperatures
- WASH policy landscape: exploring context, process, content, and actors in childcare centres in Nairobi's informal settlements.
- A Policy-Based Intervention for Physical Activity and Screen Time in Childcare Outside School Hours
BibTeX
@article{matsumoto2026demonstration,
title = {{Demonstration of a compatibility-based childcare support service using quantum annealing}},
author = {Yuuma Matsumoto and Taisei Takabayashi and Rima Sato and Rumiko Honda and Masayuki Ohzeki},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-60894-z},
url = {https://doi.org/10.1038/s41598-026-60894-z},
note = {CC BY},
}
RIS
TY - JOUR
TI - Demonstration of a compatibility-based childcare support service using quantum annealing
AU - Yuuma Matsumoto
AU - Taisei Takabayashi
AU - Rima Sato
AU - Rumiko Honda
AU - Masayuki Ohzeki
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-60894-z
UR - https://doi.org/10.1038/s41598-026-60894-z
AB - In contemporary Japan, isolated parenting has become a serious social issue, increasing psychological stress on parents and potentially affecting children’s development. Existing childcare support services tend to focus on physical assistance, while psychological support and community connections remain insufficient. To address this gap, we developed a service that connects parents with senior community members who have parenting experience, aiming to provide psychological support and foster intergenerational exchange. Achieving high-quality matching requires considering pair compatibility, balancing supporter workload, and handling scheduling constraints, which can be formulated as a combinatorial optimization problem. We represented this matching problem as a Quadratic Unconstrained Binary Optimization (QUBO) model and evaluated a quantum-annealing-based sampling pipeline against simulated annealing (SA) and the exact mixed-integer optimization solver Gurobi. In randomized benchmark instances, the QA-based pipeline generated high-quality and diverse matching candidates, particularly for larger instances; at n=15 , the mean best relative error was 1.94% for QA with post-processing and 7.45% for SA. Furthermore, a proof-of-concept field experiment conducted in Sendai City, Japan, demonstrated that the framework can generate multiple high-quality matching candidates, enabling flexible scheduling in real-world operations.
SN - 2045-2322
ER -