線形計画ソルバーと機械学習を用いた介護タクシー配車問題の最適化
Optimizing Nursing Care Taxi Dispatch Leveraging Integer Linear Programming Solvers and Machine Learning
The aging population is creating crucial demand for the transportation services at the facilities for older people. Vehicles should be assigned properly to meet users’ personal requirements and give them comfortable transportation. For example, users requiring a wheelchair need special seats.Moreover, compatibility among users is also important: if two users do not get along with each other, they should be assigned to different vehicles to reduce stress.
To address these difficulties, we formulate a new dispatch problem based on a famous combinatorial optimization problem: the Vehicle Routing Problem (VRP). The NP-hardness of VRP makes it difficult to produce high-quality solutions in a short time using exact methods, which is not ideal for real-world applications. In our research, using superior solutions generated by Integer Linear Programming (ILP) solver in advance, AI learns ideal trajectories and produces high-quality solutions in only a few seconds.
Published Papers：
- 中尾陸, 廣森聡仁, and 山口弘純. "介護送迎のための制約付き配送計画問題の定式化および機械学習を用いた解法の検討." マルチメディア, 分散, 協調とモバイルシンポジウム 2024 論文集 2024 (2024): 1714-1724.
- 中尾陸, 廣森聡仁, and 山口弘純. "スマートシティ基盤上での介護タクシー配車サービスの実現に関する検討." 第 32 回マルチメディア通信と分散処理ワークショップ論文集 (2024): 85-92.