3. iii
Applying Ant Colony Optimization to Solve Two-Stage Integrated
Production and Distribution Problem in Supply Chains
Department of Industrial Engineering and Management
National Chiao Tung University
Student:Yung-Chia Chang
Advisor:Yung-Chia Chang
Abstracts
Due to the rapid-changing market demand and highly customized product requirements,
many enterprises have adopted make-to-order or direct-order business model. In this type of
business model, enterprises are forced to lower the amount of inventory needed across their
supply chain but still have to be more responsive to customers’ requirements. Reduced
inventory creates a closer interaction between production and distribution activities and thus
increases the practical usefulness of integrated models.
We consider an integrated production and distribution problem at the individual job level
in this study. In this problem, jobs are first processed by one of a set of unrelated parallel
machines and then distributed by vehicles with limited capacity to the corresponding
customer locations. The completion time of a job is defined as the time when it is delivered to
its customer. The objective is to find a joint production and distribution schedule so that the
total weighted completion time is minimized. The complexity of the studied problem is
NP-hard. Therefore, we used an ant colony algorithm to solve this problem in order to find
near-optimal solutions in reasonable computation times. Computational analysis is performed
to evaluate the effectiveness and stability of the proposed approach. We expect our research
results to help make the study of integrated production and distribution problems more
practical and with better application values.
Key words: production and distribution, scheduling, ant colony optimization
25. 17
選擇最佳解中的部份路段來建構為來的新路徑。其中 L 為此次迭代中的最
佳路徑。
1
)()1( −
+−= Lold
ij
new
ij ατατ (2-4)
步驟四:測試停止條件
當演算法達到終止條件時便停止,否則回到步驟二繼續搜尋。
2.2.3 蟻群演算法的相關應用蟻群演算法的相關應用蟻群演算法的相關應用蟻群演算法的相關應用
蟻群演算法除了在早期應用在TSP以及前述的VRP、非等效平行機台
問題外,還被應用來解求許多其他NP-hard問題及各種組合最佳化問題如:
二次指派問題(quadratic assignment problem)、各種排程問題(scheduling
problem)、著色問題(graph coloring problem)、網路途程問題(networks
routing problem)、連續性順序問題(sequential odering problem)、最短母
字串問題(shortest common super-sequence problem)、一般分配問題(general
assignment problem)及複合背包問題(multiple knapsack problem)等,且
經過實驗證實蟻群演算法應用在這些問題上均可獲得品質良好的解(蔡志
強,2003)。以下將簡單介紹蟻群演算法應用於與本研究較為相關的排程
問題上。
2.2.3.1 蟻群演算法應用於排程問題蟻群演算法應用於排程問題蟻群演算法應用於排程問題蟻群演算法應用於排程問題
Colorni et al. (1994)首先將蟻群系統應用在排程問題上,他們為了解決
以最小完工時間為目標的零工式生產排程問題,而提出所謂的AS-JSP 演
算法,研究結果顯示當工件與機器數小於15時,此演算法所找出的解均能
達到與最佳解差距在10%以內的水準。
Stützle (1998)則研究如何將蟻群演算法應用在流程式生產(flow-shop)
排程問題,並發表了MMAS-FSP演算法,研究結果顯示此演算法可以求得
不錯的解。Bauer et al. (1999)則考慮了單機總延遲時間排程問題(single
26. 18
machine total tardiness problem),他們提出的演算法ACS-SMTTP 省略了
費洛蒙的區域更新,並加入了多種啟發函數來協助求解,如:EDD(earliest
due date)法則與MDD(modified due date)法則,而求解100個工件之問題
的測試結果顯示,ACS-SMTTP 具備求得近似最佳解的能力。
2.3 供應鏈整合之相關文獻供應鏈整合之相關文獻供應鏈整合之相關文獻供應鏈整合之相關文獻
過去已有不少作者研究供應鏈管理中的相關整合問題,Thomas and
Griffin (1996)整理了供應鏈整合的相關文獻且作深入的分析,他們將供應
鏈整合的模型分為三種類型,分別為買方-賣方協調整合(buyer-vendor
coordination) 、製造-配送整合(production-distribution coordination)以及存
貨-配送整合(inventory-distribution coordination),並指出雖然分別討論產品
製造及成品配送的文獻相當豐富,但是鮮少有同時考慮這兩個階段的模
型,原因在於這類問題較不易求解,且這兩個階段往往被緩衝的存貨所分
開並由不同的部門來管理。此外,作者也建議對於此類供應鏈整合問題應
有更多研究將重點放在作業層面而非策略層面。
Chen and Vairaktarakis (2005)也指出,大部分的製造-配送-存貨模型
都只將焦點放在管理者決策的策略層面,而較少研究細部排程的整合決
策。而且在這些模型中,製造與配送階段是經由存貨間接連結,存貨成本
也佔了相當高的比例。而這些模型已不符合現今企業採用直接銷售模式以
降低存貨成本的概念,因此必須發展更多的學術研究來建構製造與配送間
直接互動的模型,以及實務上可行的求解技巧。
Lee and Chen (2001)探討了包含運送部份的機台排程問題,並針對兩
種不同類型的問題作研究:第一類問題為半成品經由自動搬運車運送到下
一個需加工的機台,其中排程部分為雙機台的流程式生產類型,運送部分
考慮了一部或多部車輛及不同的容量限制,問題目標則是最小化最大完工
時間;第二類問題討論加工完畢後的成品運送到顧客或是倉庫,機台排程
27. 19
部分考慮了單一機台、雙機平行機台及雙機流程式生產,運送部份除了單
一機台情形下考慮多部車輛外,其餘皆只探討單一車輛的情況,問題目標
則是最小化最大完工時間或總完工時間。作者針對各種不同問題,釐清其
問題複雜度,或是提出可在多項式時間內求解的演算法。
Hall and Potts (2002)結合了供應鏈管理與排程的概念,討論供應鏈中
各種排程、批次及運送的整合問題。有別於以往的排程文獻,作者除了分
別以供應商與製造商的觀點來探討決策問題外,還考慮了這兩者的整合決
策,以求整體系統成本的最小化。在排程問題部分,作者將一個供應商或
製造商視為一個單一機台;在批次運送部份,假設車輛沒有容量上限,而
同一個顧客所下的訂單屬於同一個批次,一部車輛一次只會將成品運送到
一個顧客點,因此不會涉及車輛途程決策的問題,但必須考慮發車數量所
造成的成本。作者針對供應商、製造商及整合的觀點,在追求不同目標式
的情況下找出問題的複雜度,並為可在多項式時間內求解的問題提出動態
規劃解法。作者並以舉例方式說明藉由整合供應商及製造商的決策,在某
些情況下至少可以降低系統總成本的20%,甚至降低成為原來的一半。
Chang and Lee (2004)探討 Lee and Chen (2001)所研究的第二類問題,
其中針對機台數量及顧客區域數假設了三種情境:單一機台運送至單一顧
客區域、二台等效機台運送至單一顧客區域及單一機台運送至兩個顧客區
域。此外還加入車輛容量的限制,並考慮訂單體積的大小。Li et al. (2003)
針對 Chang and Lee (2004)之研究作了延伸,將所欲服務的顧客數延伸到多
個。他們探討單一機台排程整合單一車輛途程決策的問題,假設有多個固
定數目的顧客,以最小化訂單交貨時間的總和為目標,並在特定簡化條件
下使用動態規劃求得最佳解,作者建議未來的研究方向可以將模型延伸至
多部車輛,或是利用啟發式解法來求解任意顧客數目的問題。Chang and
Lee(2003)探討數種整合兩階段的問題,其中包括 Lee and Chen (2001)所研
究的兩類問題。作者將這兩個階段視為一個系統,利用兩種在實務上常被
68. 60
Allahverdi, A., J. Mittenthal. " Scheduling on M parallel machines subject to
random breakdowns to minimize expected mean flow time ". Naval Research
Logistics, 41 (5), pp. 677-682, 1994.
Azizoglu, M., O. Kirka (1999). Scheduling jobs on unrelated parallel machines
to minimize regular total cost functions. IIE Transactions, 31, 153–159.
Baker, B. M. and M. A. Ayechew (2003). A genetic algorithm for the vehicle
routing problem. Computers & Operations Research, 30, 787-800.
Bauer, A., B. Bullnheimer, R. F. Hartl, and C. Strauss(1999). Minimizing total
tardiness on a single machine using ant colony optimization. In Proceedings of
the 1999 Congress on Evolutionary Computation (pp. 1445-1450). Piscataway,
New Jersey: IEEE Press.
Bell, J. E., P. R. McMullen (2004). Ant colony optimization techniques for the
vehicle routing problem. Advanced Engineering Informatics, 18, 41–48.
Bruno, L. G., Coffman, E. G. and Sethi, R. (1974). Scheduling independent
tasks to reduce mean finishing time. Communications of the ACM, 17,
382–387.
Bullnheimer, B., R. F. Hartl, and C. Strauss (1997). Applying the ant system to
the vehicle routing problem. Presented at the 2nd International Conference on
Meta-heuristic-Mic97, Sophia-Antipolis, France.
69. 61
Chang, Y. C. , C. Y. Lee (2003). Logistics scheduling: analysis of two-stage
problems. Journal of System Science And Systems Engineering, 12 (4),
371-393.
Chang, Y. C. , C. Y. Lee (2004). Machine scheduling with job delivery
coordination. European Journal of Operation Research, 158 (2), 470-487 .
Chen, Z. L., G. Vairaktarakis (2005). Integrated scheduling of production and
distribution operations. Management Science, 51 (4), 614–628.
Christofides N., A. Mingozzi, and P. Toth (1981). Exact algorithm for the
vehicle routing problem, based on spanning tree and shortest path relaxations.
Mathematical Programming, 20, 255-282.
Clarke, G., J. W. Wright (1964). Scheduling of vehicles from a central depot to
a number of delivery points. Operations Research, 12, 568-581.
Colorni A., M. Dorigo, and V. Maniezzo (1991). Distributed optimization by
ant colonies. F. J. Varela and P. Bourgine (Ed.), In Proceedings of the First
European Conference on Artificial Life (pp. 134-142). Cambridge,
Massachusetts: MIP Press.
Colorni, A., M. Dorigo, V. Maniezzo, and M. Trubian (1994). Ant system for
job-shop scheduling. Belgian Journal of Operations Research, Statistics and
Computer Science (JORBEL), 34 (1), 39-53.
70. 62
Dhaenens-Flipo, C., G. Finke (2001). An integrated model for an industrial
production-distribution problem. International Journal of Production and
Economics,74 (1-3), 93-101.
Dorigo, M., L. M. Gambardella (1997). Ant colony system: A cooperative
learning approach to the traveling salesman problem. IEEE Transactions on
Evolutionary Computation, 1(1), 53-66.
Eilon, S., C .Watson-Grandy, N. Christofides (1971). Distribution management:
mathematicak modeling and practical analysis. New York: Hanfer.
Garcia, J. M., S. Lozano (2005). Production and delivery scheduling problem
with time windows. Computers & Industrial Engineering, 48, 733–742.
Gendreau, M., A. Hertz and G. Laporte (1994). A tabu search heuristic for the
routing problem. Management Science, 40, 1276-1290.
Gillett, B. E., , L. R. Miller (1974). A heuristic algorithm for the vehicle
dispatch problem. Operations Research, 22, 240-349.
Glass, C. A., C. N. Potts and P. Shade (1994). Unrelated parallel machine
scheduling using local search. Mathematical and Computer Modeling, 20 (2),
41–52.
Hall, N. G. , C. N. Potts (2003). Supply chain scheduling: batching and
delivery. Operations Research, 51, 566–584.
71. 63
Hariri, A. M. A., C. N. Potts (1991). Heuristics for scheduling unrelated
parallel machines. Computers and Operations Research, 18 (3), 323–331.
Horn, W. A. (1973). Minimizing average flow time with parallel machines.
Operations Research, 21, 846–847.
Kim, D. W., K. H. Kim, W. Jang and F. F. Chen (2002). Unrelated parallel
machine scheduling with setup times using simulated annealing. Robotics and
Computer Integrated Manufacturing, 18 (3–4), 223–231.
Kolen, A. W. J., A. H. G. R. Kan and H. W. J. M. Trienkens (1987). Vehicle
routing with time windows. Operations Research, 35 (2), 266-273.
Lee, Y.H., Bhaskaran, K., and Pinedo, M. (1997). A heuristic to minimize the
total weighted tardiness with sequence dependent setups. IIE Transactions, 29,
45-52.
Lee, C. Y., Z. L. Chen (2001). Machine scheduling with transportation
considerations. Journal of Scheduling, 4, 3–24.
Lenstra, J., A. Rinnooy Kan. (1981). Complexity of vehicle routing and
scheduling problems. Networks, 11, 221-227.
Li, C. L., G. Vairaktarakis, C. Y. Lee. (2003). Machine scheduling with
deliveries to multiple customer locations. European Journal of Operational
Research, 164, 39–51.
72. 64
Mole, R.H., S.R. Jameson (1976). A sequential route-building algorithm
employing a generalized savings criterion. Operational Research Quarterly, 27,
503-511.
Osman, I., (1993). Metastrategy simulated annealing and tabu search
algorithms for the vehicle routing problem. Annals of Operations Research, 41,
421-51
Garey R. and Johnson D. S. (1979). Computers and intractability: A guide to
the theory of NP completeness. New York: W. H. Freeman and Company.
Renaud, J., G. Laporte and F. F. Boctor (1996). A tabu search heuristics for the
multi-depot vehicle routing problems. Computer& Operations Research, 23 (3),
229-235.
Robusté, F., C. F. Daganzo, and R. Souleyrette II (1990). Implementing
Vehicle Routing Models. Transportation Research, 24B, 263-286.
Sarmiento, A. M., R. Nagi (1999). A review of integrated analysis of
production-distribution systems. IIE Transactions 31 (11), 1061-1074.
Srivastava, B. (1997). An effective heuristic for minimizing makespan on
unrelated parallel machines. Journal of the Operational Research Society, 49,
886–894.
73. 65
Stützle, T.(1998). An ant approach to the flow shop problem. In Proceedings of
the 6th European Congress on Intelligent Techniques & Soft
Computing(EUFIT'98) (pp. 1560-1564). Aachen: Verlag Mainz.
Suresh, V., C. Dipak (1994). Minimizing maximum tardiness for unrelated
parallel machines. Production Economics, 34, 223-229.
Thomas, D. J. and P. M. Griffin. (1996). Coordinated supply chain
management. European Journal of Operational Research, 94, 1-15.
Van Breedam, A. (1995) Improvement heuristics for the vehicle routing
problem based on simulated annealing. European Journal of Operational
Research, 86 (3), 480-490.
Weng, M. , J. Lu and H. Ren (2001). Unrelated parallel machine scheduling
with setup consideration and a total weighted completion time objective.
International Journal of Production Economics, 70, 215–226.
Willard J. A. G. (1989). Vehicle routing using γ-optimal tabu search, Master of
Science Dissertation, The Management School, Imperial College, London.
Wren, A., A. Holliday (1972). Computer scheduling of vehicles form one or
more depots to a number of delivery points. Operational Research Quarterly,
23, 333-344.