都市や観光地における混雑状況を考慮した多数ユーザ同時巡回スケジューリング手法  [in Japanese] Route Scheduling Technique for Massive Users in City Section and Sightseeing Area Based on Congestion of Roads and Spots  [in Japanese]

Access this Article

Search this Article

Abstract

都市部や観光地における移動経路,訪問施設の混雑状況を緩和するため,これまで,ユーザ間の情報共有による重複しない移動経路の選択法や,直近の混雑情報から混雑していない施設を訪問するための手法が提案されている.本論文では,各移動経路,各目的地における混雑を同時に考慮した,複数ユーザ同時巡回スケジューリング手法を提案する.提案手法では,各ユーザの希望巡回目的地リストを入力とし,(i) すべてのユーザが希望した目的地を同時に巡回したと仮定したシミュレーションを行うことで,時刻ごとの各道路,各施設の混雑状況を予測したうえで,(ii) 予測混雑下においてユーザの要望をできるだけ満足するよう,ユーザが巡回する目的地の調整を繰り返すことでスケジュールを算出する.提案手法を計算機上に実装し,都市部道路網を模した地図を用いて評価実験を行った結果,20,000のユーザについて,8分程度と実用化可能な時間でスケジューリングでき,ユーザが直近の混雑情報を用いて行動した場合と比べて,提案方式ではより高い満足度(ユーザが希望どおりに巡回できたかを示す指標)を達成できることを確認した.また,提示されたスケジュールを無視して独自の判断で行動するユーザが出現する場合を想定した条件設定の下においても実験を行い,提案手法が有効に機能することを確認した.So far, a number of scheduling methods have been proposed to find (i) a noncongested route by sharing route information among users, or (ii) a schedule to alleviate congestion at specific facilities based on the latest congestion information. In this paper, we propose a method for finding schedules for massive users by predicting congestion of routes and destinations at each point of time. In our method, we collect all users' information of visiting destinations and perform traffic simulation assuming that the users visit their intended destinations. Using the simulation result, our method adjusts each user's provisional schedule by changing visiting order of destinations and reducing the number of visiting destinations, while keeping the user's satisfaction as high as possible. We have implemented the proposed method and evaluated the method by performing simulations for a large-scale road network with more than 20,000 users. As a result, we confirmed that the computation time is about eight minutes, which is reasonable for practical use. Our experimental results also show that, (1) the proposed method achieves higher user satisfaction degree (the sum of importance degrees specified for the visited destinations) than the case when users visit each destination by using the latest congestion information of roads and facilities, (2) users have reasonable incentive to follow the output.

So far, a number of scheduling methods have been proposed to find (i) a noncongested route by sharing route information among users, or (ii) a schedule to alleviate congestion at specific facilities based on the latest congestion information. In this paper, we propose a method for finding schedules for massive users by predicting congestion of routes and destinations at each point of time. In our method, we collect all users' information of visiting destinations and perform traffic simulation assuming that the users visit their intended destinations. Using the simulation result, our method adjusts each user's provisional schedule by changing visiting order of destinations and reducing the number of visiting destinations, while keeping the user's satisfaction as high as possible. We have implemented the proposed method and evaluated the method by performing simulations for a large-scale road network with more than 20,000 users. As a result, we confirmed that the computation time is about eight minutes, which is reasonable for practical use. Our experimental results also show that, (1) the proposed method achieves higher user satisfaction degree (the sum of importance degrees specified for the visited destinations) than the case when users visit each destination by using the latest congestion information of roads and facilities, (2) users have reasonable incentive to follow the output.

Journal

  • 情報処理学会論文誌

    情報処理学会論文誌 51(3), 885-898, 2010-03-15

    一般社団法人情報処理学会

Codes

  • NII Article ID (NAID)
    110007970692
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • Data Source
    NII-ELS  IR  IPSJ 
Page Top