當地震或是其他災害發生時，災後的物資會成為稀缺資源，政府作為決策者，
迅速且有效地決策與規劃物資的派送路線，可以降低災民的焦慮、民眾的傷亡以
及災後秩序的混亂。所以減災、備災，以及災後的應變是災害管理的重要議題。
災害具有突發、不可預料以及重傷害的特性，使災後的運輸網路有大量不確定性
因子，例如：災後道路的失效以及物資需求，增添物資分配問題的複雜性，造成
管理供應鏈困難。
本研究提出一個二階段隨機混整數規劃模型，建構一個在時間限制下的物資
配送架構，以求解物資集散地的位置與開設數、所需車輛數、車輛路徑以及其運
送物資量的最佳決策問題，以利用模擬最佳化方法求解最小化成本。此外，亦設
計一組合式演算法來求解。演算法以樣本平均近似法為基礎，透過隨機抽樣的方
式將隨機參數轉換成大量情境值，將隨機混整數規劃模型轉化為可求解的確定性
混整數規劃模型，並且以位址快速篩選法以及分配巢狀分割法來解決此問題。最
後，本研究使用國家災害防救科技中心（National Science and Technology Center
for Disaster Reduction, NCDR）之資料，進行實證研究，除了尋找出影響成本的重
要因子外，也分析在不同情境下車輛配置數與開設位址的最佳決策，提供決策者
災後物資分配的實務參考，使本研究在降低災民脆弱度上有所貢獻。

During times of earthquakes or other disasters, the availability of post-disaster supplies becomes scarce, leading to anxiety among victims, casualties, and chaotic conditions in the aftermath. To alleviate these concerns, the government plays a crucial role as the decision-maker by swiftly and effectively planning the delivery routes for supplies. However, the unpredictable and severe nature of disasters introduces significant uncertainty into the post-disaster transport network. This uncertainty encompasses road failures and fluctuating supply demands, thereby increasing the complexity of managing the supply chain distribution problem. In this study, we propose a two-stage stochastic mixed-integer planning model aimed at developing a time-constrained distribution structure. The model aims to solve the optimal decision problem concerning the location and quantity of distribution sites, the required number of vehicles, the vehicle routes, and the number of supplies to be delivered while minimizing costs. Also, an algorithm is developed, which employs a sample-average approximation method with utilizing rapid location procedure and Nested Partitions-based Allocation Procedure.

Overall, this study contributes to the field of disaster management by presenting a comprehensive planning model and algorithm for post-disaster supply chain distribution. We conduct an empirical study using data from the National Science and Technology Center for Disaster Reduction (NCDR). The empirical analysis based on this real data helps identify important factors and informs optimal decisions on vehicle allocation and site selection, providing practical guidance for post-disaster material allocation. By addressing the challenges posed by uncertain post-disaster conditions, we aim to enhance the efficiency of resource allocation in the aftermath of disasters.

李洋寧、劉淑燕、李沁妍、吳佳容、鄧敏政、柯孝勳、李中生 (2014) 大臺北地 區大規模地震衝擊情境分析報告 II ：道路系統、水電設施、重要設施、情境綜整。NCDR 102-T15。國家災害防救科技中心。
Boonmee, C., M. Arimura, T. Asada (2017). Facility location optimization model for emergency humanitarian logistics, International Journal of Disaster Risk Reduction 24, 485-498.
Bosse, A., M. W. Ulmer, E. Manni, D. C. Mattfeld (2023). Dynamic priority rules for combining on-demand passenger transportation and transportation of goods, European Journal of Operational Research 309(1), 399-408.
Coppola, D.P. (2011). Introduction to International Disaster Management, chap 4 - chap 7, Boston: Butterworth-Heinemann, Print, 209-423.
Chang, K.H. and C. C. Huang (2023). An interactive two-stage framework for simultaneous machine selection and buffer allocation, IEEE Transactions on Automation Science and Engineering 20(1), 675-689.
Chen, C.H., L.H. Lee (2010). Stochastic simulation optimization - An optimal computing budget allocation. World Scientific Publishing Co., Inc.
Dönmez, Z., B. Y. Kara, Ö. Karsu, F. Saldanha-da-Gama (2021). Humanitarian facility location under uncertainty: Critical review and future prospects, Omega 102, 102393.
Erbeyoğlu, G., Ü. Bilge (2020). A robust disaster preparedness model for effective and fair disaster response, European Journal of Operational Research 280(2), 479-494.
FEMA (2022). Hazus Earthquake Model Technical Manual, Hazus 5.1, 203-205.
Fenton, G., S. Hale, S. Lawry-White, et al. (2019). Evaluation of humanitarian logistics within EU civil protection and humanitarian action, 2013-2017 : final report, chap 2.
Fu, M.C. (2002). Optimization for simulation: theory vs. practice, INFORMS Journal on Computing 14(3), 192-215.
Gaddam, H.K., K.R. Rao (2019). Speed–density functional relationship for heterogeneous traffic data: a statistical and theoretical investigation. Journal of Modern Transportation 27, 61–74.
Healey, J. F., C. Donoghue, (2020). Statistics: A Tool for Social Research and Data Analysis. America: Cengage Learning.
Hoyos, M.C., R.S. Morales, R. Akhavan-Tabatabae. (2015). OR models with stochastic components in disaster operations management: a literature survey, Computers & Industrial Engineering 82, 183-197.
Hu, H., S. Guo, Y. Qin, W. Lin (2023). Two-stage stochastic programming model and algorithm for mitigating supply disruption risk on aircraft manufacturing supply chain network design, Computers & Industrial Engineering 175, 108880.
Kubat, M. (2017). An Introduction to Machine Learning, second edition, chap 16 Springer International Publishing, https://doi.org/10.1007/978-3-319-63913-0.
Li, Y., F. Saldanha-da-Gama, M. Liu, Z. Yang (2023). A risk-averse two-stage stochastic programming model for a joint multi-item capacitated line balancing and lot-sizing problem, European Journal of Operational Research 304(1), 353-365.
Mahtab, Z., A. Azeem, S. M. Ali, S. K. Paul, A. M. Fathollahi-Fard (2022). Multi-objective robust-stochastic optimisation of relief goods distribution under uncertainty: a real-life case study, International Journal of Systems Science: Operations & Logistics 9(2), 241-262.
May, A.D., H.E Keller (1967). Non-integer car-following models. Highway Research Record 199(1), 19-32.
Nahas, N., Nourelfath, M. (2015). Buffer allocation, machine selection and preventive maintenance optimization in unreliable production lines. 2015 International Conference on Industrial Engineering and Systems Management (IESM).
Nikbakhsh, E., R.Z. Farahani (2011). Humanitarian logistics planning in disaster relief operations, Logistics Operations and Management: Concept and Models 15, 291-332.
Rennemo, S.J., K.F. Rø, L. M. Hvattum, G. Tirado (2014). A three-stage stochastic facility routing model for disaster response planning, Transportation Research Part E: Logistics and Transportation Review 62, 116-135.
Rezaei-Malek, M., R. Tavakkoli-Moghaddam, B. Zahiri, A. Bozorgi-Amiri (2016). An interactive approach for designing a robust disaster relief logistics network with perishable commodities, Computers & Industrial Engineering 94, 201-215.
Shi, L., S. Ólafsson (2000). Nested partitions method for global optimization, Operations Research 48(3), 390-407
Shokr, I., F. Jolai, A. Bozorgi-Amiri (2022). A collaborative humanitarian relief chain design for disaster response, Computers & Industrial Engineering 172, Part A, 108643
Tsai, S. C., (2013). Rapid screening procedures for zero-one optimization via simulation, INFORMS Journal on Computing 25(2), 193-393.
Wallace, W.A., F.D. Balogh (1985). Decision support systems for disaster management, Public Administration Review 45, 134-140.
Yang,Y., Y. Yin, D. Wang, J. Ignatius, T.C.E. Cheng, L. Dhamotharan (2023). Distributionally robust multi-period location-allocation with multiple resources and capacity levels in humanitarian logistics, European Journal of Operational Research 305(3), 1042-1062.
Yen, J.Y. (1971). Finding the k shortest loopless paths in a network, Management Science 17(11), 712-716