近年來，公共自行車共享系統已在多個國家的都會區營運，也帶動了新型態的物流研究議題。目前此類系統在營運期間最棘手的挑戰為「無車可借」或「無位可還」的供需失衡問題，現行作法大多雇用卡車，自較多車的租借站移車，再將這些車載至較少車的租借站補車，然而少量的卡車無法滿足大量租借站同時缺車或缺位的運補需求。因此，本研究提出「群眾運補」的概念，在給予適當獎勵的情況下，由系統招募志願接受運補任務的群眾，這些群眾將依系統指示於特定時刻至指定的租借站租車，再騎至指定的租借站還車，如此即可同時在多個租借站間靠群眾運補，不會受到少數卡車所在位置及容量的限制，理應比現行使用少數卡車的運補方式更為有效。
本研究分成三個階段：在第一階段的資料分析中，我們以台北市YouBike系統的公開資料為基礎，藉由分析該資料找出租還需求的變化趨勢（譬如各站每日租借量隨時間的波動方式，以及其可能還車的訖點分佈方式等），並實作多種資料視覺化工具以呈現分析結果；第二階段探討群眾運補的數學理論基礎，建構一個理想的最適車數整數規劃模式，將一天分成多期，以計算出各租借站於各期期末的最佳庫存之自行車數量，該數值可被視為該站該時刻的最適車數，再以即時租借資料與機器學習中的隨機森林預測模型來預測需求，建構一個簡化的群眾運補數學模式，以計算當期最佳的群眾號召方式；第三階段則實作一個模擬系統，模擬以群眾（或卡車）運補的情境下，如何配置這些騎乘者（或卡車）的起訖運補方式，以減少最多的預估缺車與缺位總量；經多次模擬測試，證實群眾運補的確比傳統卡車運補有更高的效益，尤其在尖峰時段更能大幅減少缺車與缺位發生的機會，可明顯地改善服務品質。

The public bike sharing system is a popular shared economy application. Its success depends on the availability of bikes and empty racks in rental sites spread out a metropolitan area with sufficient density. Its major management challenge is to reposition bikes between rental sites to prevent full or empty rental sites. To this end, most systems hire trucks to move bikes between rental sites, yet this repositioning scheme is ineffective because the number of trucks are usually much fewer than rental sites. Here we proposed a crowdsourced repositioning scheme, where voluntary riders are hired to ride specific Origin-Destination trips. Receiving satisfactory bonus in return, the voluntary crowd can move bikes at all rental sites simultaneously with less costs.
We conduct in-depth analysis on historical rental data to identify important factors and features affecting demands. Some visualization tools are designed to provide better understanding to the demand dynamics. The random forest algorithm is employed to predict the expected incoming bike rentals and returns. Setting every 30 mins as a period, we propose an Ideal Inventory Model (IIM) using integer program, and calculate optimal bike inventories for all rental sites and periods. To deal with on-line repositioning, we propose the Voluntary Rider Flow Model for period t (VRFM(t)) that calculates voluntary rider flows by a greedy algorithm. The results of our tests by simulation indicate our crowdsourced repositioning strategy does provide better service quality than the truck repositioning strategy.

周佰賢. (2015) 考慮需求變化狀況及增設臨停區之公共自行車共享系統租借站分群與車輛調度策略研究. 工業與資訊管理學系碩士論文，國立成功大學
洪菁蓬. (2011) 公共自行車共享系統之最佳租借站位址設置及車輛運補策略之研究. 工業與資訊管理學系碩士論文，國立成功大學
張立蓁. (2010) 都會區公共自行車共享系統之設計與營運方式研究. 工業與資訊管理學系碩士論文，國立成功大學
廖敏婷. (2012) 考慮需求比例及暫時人力配置之公共自行車共享系統管理策略研究. 工業與資訊管理學系碩士論文，國立成功大學
Bargar, A., Gupta, A., Gupta, S., and Ma, D. "Interactive Visual Analytics for Multi-City Bike share Data Analysis." The 3rd International Workshop on Urban Computing (UrbComp 2014), New York, USA. (2014).
Breiman, L. "Random forests." Machine learning 45.1 (2001): 5-32.
Cagliero, L., Cerquitelli, T., Chiusano, S., Garza, P., and Xiao, X. "Predicting critical conditions in bicycle sharing systems. " Computing, (2016): 1-19.
Contardo, C., Morency, C., and Rousseau, L. M. "Balancing a dynamic public bike-sharing system." Centre Interuniversitaire de Recherche sur les Réseaux d'Entreprise, la Logistique et le Transport, Vol. 4. (2012).
Conway , M. W. "Predicting the Popularity of Bicycle Sharing Stations: An Accessibility-Based Approach Using Linear Regression and Random Forests." (2014).
DeMaio, P. "Bike-sharing: History, impacts, models of provision, and future." Journal of Public Transportation 12.4 (2009): 3.
Froehlich, J., Neumann, J., and Oliver, N. "Sensing and Predicting the Pulse of the City through Shared Bicycling." International Joint Conference on Artificial Intelligence. Vol. 9. (2009): 1420-1426.
Gebhart, K, and Noland, R. B. "The impact of weather conditions on bikeshare trips in Washington, DC." Transportation 41.6 (2014): 1205-1225.
Kaltenbrunner, A., Meza, R., Grivolla, J., Codina, J., and Banchs, R. "Urban cycles and mobility patterns: Exploring and predicting trends in a bicycle-based public transport system". Pervasive and Mobile Computing, 6(4), (2010): 455-466.
Kaspi, M., Raviv, T., Tzur, M., and Galili, H. "Regulating vehicle sharing systems through parking reservation policies: Analysis and performance bounds." European Journal of Operational Research, 251(3), (2016): 969-987.
Montoliu, R. "Discovering mobility patterns on bicycle-based public transportation system by using probabilistic topic models." Ambient Intelligence-Software and Applications. Springer Berlin Heidelberg, (2012): 145-153.
O’brien, O., Cheshire, J., and Batty, M. "Mining bicycle sharing data for generating insights into sustainable transport systems." Journal of Transport Geography 34 (2014): 262-273.
O’Mahony, E. D. "Smarter Tools For (Citi) Bike Sharing. " PhD Thesis. Cornell University. (2015).
Raviv, T., Tzur, M., and Forma, I. A. "Static repositioning in a bike-sharing system: models and solution approaches." EURO Journal on Transportation and Logistics 2.3 (2013): 187-229.
Rixey, R. "Station-Level Forecasting of Bikesharing Ridership: Station Network Effects in Three US Systems." Transportation Research Record: Journal of the Transportation Research Board 2387 (2013): 46-55.
Sarkar, A., Lathia, N., and Mascolo, C. "Comparing cities’ cycling patterns using online shared bicycle maps." Transportation 42.4 (2015): 541-559.
Vogel, P., and Mattfeld, D. C. "Strategic and operational planning of bike-sharing systems by data mining–a case study." International Conference on Computational Logistics. Springer Berlin Heidelberg. (2011)
Vogel, P., Greiser, T., and Mattfeld, D. C. "Understanding bike-sharing systems using data mining: Exploring activity patterns." Procedia-Social and Behavioral Sciences 20 (2011): 514-523.
Vogel, P., Saavedra, B. A. N., and Mattfeld, D. C. "A hybrid metaheuristic to solve the resource allocation problem in bike sharing systems. " International Workshop on Hybrid Metaheuristics. Springer International Publishing. (2014): 16-29.
Yang, Z., Hu, J., Shu, Y., Cheng, P., Chen, J., and Moscibroda, T. "Mobility Modeling and Prediction in Bike-Sharing Systems". MobiSys '16 Proceedings of the 14th Annual International Conference on Mobile Systems, Applications, and Services (2016): 165-178
Yoon, J. W., Pinelli, F., and Calabrese, F. "Cityride: a predictive bike sharing journey advisor." 2012 IEEE 13th International Conference on Mobile Data Management. IEEE, (2012): 306-311