零擔貨運業者乃未滿整櫃貨物載運者，因此有必要透過集合組裝以達整櫃運送之效益。而近年來業者為了改善效率，逐漸將營運網路轉換成為業界常採用之軸輻式營運網路，以降低營運成本。
路線貨運業者內部營運計畫常同時探討貨物路徑規劃、貨櫃型態和車輛指派、與司機排班規劃。相關文獻探討業者營運皆以內部營運作業為一天一次的載運週期為主，並利用一長期收集而來的載運貨量平均值，假設為研究已知的貨物運輸需求量。本研究以為在載運網路中，站所在一天之運作週期中僅做一次的發送及到著的作業週期，則會使有許多時間為站所及車輛閒置，如此並不具效率，且假設歷史貨物載運量作為已知之貨物需求，更為不合理。本研究提出多時段載運，為使路線貨運業內部營運網路在一天之中可進行兩次的作業週期，即於可行的時段內可多次發送，並利用推測性規劃的觀念來解決需求不確定性為未知的特性。
本研究嘗試由問題特性歸納與分析後，考慮多時段載運網路及需求不確定之特性，必須考慮時間與決策之問題，故本研究藉時空架構構建動態網路以考量於不同的決策點，必須決定之貨物路徑與車輛路線，並進而構建數學模式以及發展可行之求解演算法進行求解，最後並進行實例研究測試，以瞭解路線貨運業多時段載運之可行性。

The time-definite freight delivery common carriers pickup, consolidate and deliver small shipments. The pure hub-and-spoke network consolidates partial loads with the result of lower operating cost. As the result, it has become the most common line-haul operations configuration for the freight delivery industry. To improve the operational efficiency, the carriers in Taiwan gradually adjusted, re-configured and transformed their line-haul operations in pure hub-and spoke networks.
The line-haul operations network planning of time-definite freight delivery common carriers includes the determination of freight paths, trailer assignment and balance, and feeder schedules to guide daily operations. Most research in the subject of operational planning limited to one daily pickup and delivery operation. Furthermore, the assumed the demand was an average of long-term historical freight demand. In practice, the demand fluctuates. In addition, multiple-frequency may potentially reduce the fleet size and overall operating cost. Thus, in this research we studied the economic effects on multiple-frequency line-haul operations under the uncertain demand. Under the uncertain demand, the multiple-stage load planning problem is to determine the freight paths and fleet size with their routes so that the sum of scheduled and expected recourse costs is minimized. The problem is a multiple-stage stochastic and integer program.
In this research, we constructed a space-time capacitated directed network for the multiple-stage line-haul operations. Subsequently, we developed a branch and bound algorithm with embedded scenario aggregation to determine the optimal operations plan. The branches were determined by scenario aggregation. For numerical experiment, we selected a set of small networks of the third largest common carrier in Taiwan. The results showed a lower operating cost than the conventionally used deterministic load plan.

1.林正章，「多元商品流量問是在唯一路徑限制下的正確演算法」，中華民國運輸學會第十屆論文研討會論文集，民國84年，頁303-310。
2.林正章，「貨物運輸網路研究之回顧與展望」，中華民國第四屆運研討會，民國88年，頁140-160。
3.林正章、劉志遠，「路線貨運業貨物整體運輸網路設計之研究」，中華民國第二屆運輸網路研討會論文集，民國86年11月，頁431-440。
4.林正章、許瓊文，「當日快遞路運路網設計之研究」，運輸學刊，第十一卷，第二期，民國八十八年，頁61-86。
5.林正章、黃冠翔，「懲罰函數法應用於即時性貨物排程問題之研究」，運輸計劃季刊，28(3)，民國88年，頁421-450。
6.林正章，「路線貨運業單一路徑限制之貨物排程規劃問題」，運輸計劃季刊，29(1)，民國89年，頁1-32。
7.吳永嘉，「路線運輸業多頻率載運規劃問題之研究」，國立成功大學交通管理研究所碩士論文，民國88年6月。
8.高玉明，「路線貨運業貨排程問題之研究」，國立成功大學交通管理研究所碩士論文，民國85年6月。
9.張文正，「零擔汽車貨運業管理資訊系統建立之研究」，碩士論文，淡紅大學管理科學研究所，民國72年。
10.陳春益、林正章、高玉明，「路線貨運業貨物排程問題之研究」，運輸計劃季刊，26(2)，民國86年，頁327-352。
11.劉志遠，「路線貨運業貨物運輸網路整體設計之研究」，國立成功大學交通管理研究所碩士論文，民國86年6月。
12.蔡焦蘭，「建立貨物排程決策支援系統之研究」，國立成功大學交通管理研究所碩士論文，民國86年6月。
13.Akyilmaz, M. O., “An Algorithmic Framework for Routing LTL Shipments”, Journal of the Operational Research Society, 45(5), 1994, pp.529-538.
14.Current, J. R.(1988)The design of a hierarchical transportation network with transshipment facilities. Transportation Science, 22(4), pp.270-277.
15.Crainic T. G. and Laporte G., “Invited Review Planning models for freight transportation”, European Journal of Operational Research, 97, pp.409-438.
16.Eckstein, Jonathan and Sheffi, Y., “Optimization of Group Line-Haul Operations for Motor Carriers Using Twin Trailers”, Transportation Research Record, 1120,1987, pp.12-23.
17.Goden S. and R. de Nufville, “Design of air transportation networks”, Transportation Research, 7, 1973, pp.207-222.
18.Leung, J. M. Y., Magnanti, T. L. and Singhal, V., “Routing in Point-Point Delivery Formulations and Solution Heuristics”, Transportation Science, 24(4), 1990, pp. 245-260.
19.Lin, Cheng-Chang, “The Load Planning of Time-Definite Freight Delivery Common Carriers”, Transportation Science, 20(2), 1986, pp. 92-106.
20.Lin, Cheng-Chang, “Load Planning at UPS”, ORSA-TIMS October National Conference, Philadelphia, PA, 1990.
21.Lin, Cheng-Chang, “The fright routing problem of time definite freight delivery common carriers.” Transportation Research B, 35(6), 2001, pp.525-547.
22.Peter Kall and Stein W. Wallance, Stochastic Programming, John Wiley & Sons, 1994.
23.Powell, W. B. and Koskosidis, I. A., “Shipment Routing Algorithms With Tree Constraints”, Transportation Science, 26(3), 1992, pp. 230-245.
24.Powell, W. B. and Sheffi, Y., “Design and Implement of an Interactive Optimization System for Network Design in the Motor Carrier Industry”, Operation Research, 37(1), 1989, pp. 12-29.