Development of a global logistics system that can respond to a rapidly changing society

2018 Fiscal Year Final Research Report

• Project/Area Number: 26350417
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Social systems engineering/Safety system
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Nakata Kazuhide 東京工業大学, 工学院, 准教授 (00312984)
• Project Period (FY): 2014-04-01 – 2019-03-31
• Keywords: 物流 / オペレーションズ・リサーチ / 最適化 / 機械学習 / メタヒューリスティック

Outline of Final Research Achievements

In order to conduct efficient logistics in a rapidly changing social environment, it is necessary to deal with uncertain factors. For this reason, we have researched methods for predicting the near future to the past data as accurately as posible using machine learning methods. Also, in order to obtain a satisfactory solution in a realistic time, we have developed a practical optimization algorithm that calculates an approximate solution stably. Using these research results, we have worked on case studies and developed efficient modeling and optimization algorithms for various real problems.
We published these results as 15 peer-reviewed articles and 9 non-peer-reviewed articles. We also made presentations at 38 domestic and international conferences, including 6 invited talks.

Free Research Field

オペレーションズ・リサーチ

Academic Significance and Societal Importance of the Research Achievements

物流の最適化において、これまでは学術的側面からの研究が主流であり、実用性には程遠い研究が多かった。本研究では実務家との協働によって現実の複雑な状況を十分に考慮し、さらに最適化と機械学習の最新技術を融合させることで、輸送ルートの決定など１週間から数ヶ月のスパンを想定するタクティカルレベル、並びに日々の輸送計画について検討するオペレーションレベルのおいて、現代の目まぐるしく変化する消費者の嗜好、予測の難しい突発的な大規模災害などに対して、効率的・合理的な物流計画が立案可能となっており実用性は高い。