2022 Fiscal Year Final Research Report
Approximate Optimization Algorithms with Theoretical Rationales in Markov Decision Processes
Project/Area Number |
19K04904
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 25010:Social systems engineering-related
|
Research Institution | Nagoya Institute of Technology |
Principal Investigator |
Nakade Koichi 名古屋工業大学, 工学(系)研究科(研究院), 教授 (50207825)
|
Project Period (FY) |
2019-04-01 – 2023-03-31
|
Keywords | マルコフ決定過程 / 最適化 / 数値解析 / アルゴリズム |
Outline of Final Research Achievements |
The appropriate algorithm for various problems including the production systems to optimize the system is hard to apply. In addition, it is difficult to collect many real data to evaluate policies appropriately and optimize the system. We developed the near-optimal policy for several problems considering their qualitive and quantitative properties. The optimal allocation of customers to servers in the queueing systems, the optimal order policies in stores with perishable goods, the optimal orders and production policies in a supply chain, optimal preventive maintenance policies in offshore wind turbines with seasonality, and optimal model in a green supply chain are discussed.
|
Free Research Field |
経営工学,オペレーションズリサーチ
|
Academic Significance and Societal Importance of the Research Achievements |
本研究では,サプライチェインをはじめとする様々な生産,物流,販売等のシステムをマルコフ決定過程で定式化し,それぞれのシステムがもつ特性に適切な(近似)最適化アルゴリズムを適用して,ある程度大きな問題に対する優れた決定政策を求め,また,そのシステムの近似最適政策が持つ性質を導くことができた.この性質を知ることにより,大規模で数値的に計算が困難な問題においてもどのような方針で運用すれば良いかを得るための知見を売ることができる.
|