Adaptive modal shift of waste-based biomass reduction logistics by sensor coordination

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K12537
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Web informatics, Service informatics
• Research Institution: Chiba University
• Principal Investigator: Arai Sachiyo 千葉大学, 大学院工学研究院, 教授 (10372575)
• Co-Investigator(Kenkyū-buntansha): 和嶋 隆昌 千葉大学, 大学院工学研究院, 准教授 (00380808) ; 矢入 郁子 上智大学, 理工学部, 准教授 (10358880)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: ロジスティクス最適化 / 還元物流 / センサネットワーク / マルチエージェントシステム

Outline of Final Research Achievements

This issue has been promoted for the purpose of optimizing the "recycling technology" and "reduction logistics connecting a set of facilities from generation to use" for the utilization of waste-based biomass. We investigated the differences in the number of recycling facilities and the processing capacity of sampled municipalities and showed the predictability of the generation patterns based on the type of waste. In the recent days, it is necessary to distribute sensors to estimate the amount of waste. Though it does not pay for price of the sensor, the improvement of the recycling technology, and the improvement of consideration to the environmental problem makes it possible practical usage.
In particular, we constructed a prototype of the management system which can select a delivery target from the relation between (1) the amount of generation, the management of location, and (2) The waste amount and the processing facility ability by cooperation between smart devices.

Free Research Field

マルチエージェント学習

Academic Significance and Societal Importance of the Research Achievements

廃棄物系バイオマスのエネルギー利用技術は，既に実用化レベルにあるにもかかわらず，運用に至っていない．この学術的背景の下，廃棄物の収集や保管場所の確保，いわゆる還元物流がボトルネックである点に着目した．一方，社会的には， 2020年の東京オリンピック開催で予想される食品廃棄物の急増，さらに安価で高性能なセンサ技術の発達を背景に，効果的な還元物流モデルを示すことの意義は大きい．モーダルシフトは，一般に車から鉄道，船舶へのシフトを指すが，ここでは廃棄物発生量，それを収容可能な保管・処理施設と交通手段，目的地の選定を含む柔軟な配送シフトをマルチエージェントモデルの学習によって獲得する方法を示した．