自行車租賃佈署暨調度最佳之化之研究

Decision models for deploying rental bikes under stochastic demands.

張勻威(Yun-Wei Chang)

指導教授 ： 顏上堯

國立中央大學/工學院/土木工程學系/碩士(2011年)

摘要

近年來，人們環保意識抬頭，「綠色運輸」之概念因應而生；為了提倡節能減碳，並達到永續發展的目的，提供民眾公共自行車租賃服務的自行車租賃業開始蓬勃發展。然而在台灣，自行車租賃系統的佈署仍停留在以人工經驗的方式進行，缺乏系統性的整體考量，常導致資源浪費。當面對「甲地租乙地還」更為複雜之問題時，其求解效率將大幅下降且僅可求得一可行解。因此，本研究針對「甲租甲還」與「甲租乙還」之租賃行為，考量確定性需求與隨機性需求之影響，建構一自行車佈署營運模式，期能提供一有效的規劃輔助工具，以幫助決策者有效地規劃自行車佈署與營運相關問題。
再者，以往文獻尚未發現有提出有效的模式以解決自行車租賃系統佈署與營運相關問題。因此，本研究將站在系統規劃者之立場，考量實務的營運目標與相關的限制條件，並以網路流動技巧，依照不同情況構建數個確定性及隨機性租賃自行車系統佈署模式。所有模式均為一整數規劃問題，本研究配合CPLEX數學規劃軟體求解模式。為評估模式的實用績效，本研究以台北縣市公共自行車租賃系統為例進行測試與分析。測試結果良好，最後提出結論與建議。

關鍵字

自行車 ； 數學規劃 ； 佈署 ； 隨機 ； 時空網路

並列摘要

Recent years, more and more people are caring about environment protection. Therefore, it brings the concept of the “Sustainable Transport” to our life. For encouraging people to lower our carbon footprint, conserve energy, and accomplish the goal of sustainable development, there are many rental bicycle services being developed. However, currently in Taiwan, rental bicycle service schedules still planed by the decision center staffs with experience. Without a systematic optimization analysis, and often result in waste resourses. While facing with more complicated problem like “rent bicycle at A station, and return it at B station”, such a manual approach is considered to be less than efficient, and may possibly result in an inferior feasible solution. As a result, focusing on “rent and return bicycle at the same station”and“rent bicycle at A station, and return it at B station”, we constructed some sure and stochastic rental bicycle models that considers the influence of sure and stochastic demand. The matching model is expected to be an effective tool for the planner to solve rental bicycle disposition and operation problems.
Moreover, little literature that proposed effective models for solving problems which relate with rental bicycle system disposition. Therefore, in this study we put ourselves in rental bicycle station proprietor position, consider operation goals in reality and constraint conditions, and employ time-space network techniques, according to different circumstances to construct some rental bicycle models. All decision models are formulated as an integer multiple problem and is solved using a mathematical programming solver. To evaluate models in practice we performed a case study based on the operating data from Taipei public rental bicycle system. The results show the model could be useful. Finally, conclusions and suggestions are given.

並列關鍵字

Mthematical programming ； Time-space network ； Disposition ； Stochastic ； Bicycle

參考文獻

9. 陳春益、邱明琦，「貨櫃航線網路設計模式之研究」，運輸計劃季刊，第三十一卷，第二期，第267-298頁(2002)。

10. 盧華安、徐育彰，「定期貨櫃航線選擇與船隊佈署之研究」，運輸計劃季刊，第三十卷，第三期，第577- 601頁(2001)。

11. 顏上堯、杜宇平、陳怡妃 (2004)，「因應臨時事件機場共用櫃檯即時指派之研究」，運輸計劃季刊，第三十三卷，第一期，第59- 81頁。

12. 顏上堯、翁綵穗，「季節轉換間緩衝期飛航排程之研究」，運輸計劃季刊，第三十卷，第四期，第891- 922頁(2001)。

3. 林士鈞，「定期貨櫃運輸船舶排程暨船期表建立之研究」，碩士論文，國立中央大學土木工程學系(2004)。