請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/102206完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 陳俊杉 | zh_TW |
| dc.contributor.advisor | Chuin-Shan Chen | en |
| dc.contributor.author | 陳羿璇 | zh_TW |
| dc.contributor.author | Yi-Syuan Chen | en |
| dc.date.accessioned | 2026-04-08T16:17:26Z | - |
| dc.date.available | 2026-04-09 | - |
| dc.date.copyright | 2026-04-08 | - |
| dc.date.issued | 2026 | - |
| dc.date.submitted | 2026-03-23 | - |
| dc.identifier.citation | 1. Aringhieri, R., et al., Emergency medical services and beyond: Addressing new challenges through a wide literature review. Computers & Operations Research, 2017. 78: p. 349–368.
2. Bélanger, V., A. Ruiz, and P. Soriano, Recent optimization models and trends in location, relocation, and dispatching of emergency medical vehicles. European Journal of Operational Research, 2019. 272(1): p. 1–23. 3. Brotcorne, L., G. Laporte, and F. Semet, Ambulance location and relocation models. European Journal of Operational Research, 2003. 147(3): p. 451–463. 4. Olasveengen, T.M., et al., European Resuscitation Council Guidelines 2021: Basic Life Support. Resuscitation, 2021. 161: p. 98–114. 5. Panchal, A.R., et al., Part 3: adult basic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation, 2020. 142(16_Suppl_2): p. S366–S468. 6. Pons, P.T., et al., Paramedic response time: does it affect patient survival? Academic Emergency Medicine, 2005. 12(7): p. 594–600. 7. Soar, J., et al., European Resuscitation Council Guidelines 2021: Adult advanced life support. Resuscitation, 2021. 161: p. 115–151. 8. Chen, A.Y., et al., Demand Forecast Using Data Analytics for the Preallocation of Ambulances. IEEE Journal of Biomedical and Health Informatics, 2016. 20(4): p. 1178–1187. 9. Lin, A.X., et al. Leveraging Machine Learning Techniques and Engineering of Multi-Nature Features for National Daily Regional Ambulance Demand Prediction. International Journal of Environmental Research and Public Health, 2020. 17, 4179 DOI: 10.3390/ijerph17114179. 10. Martin, R.J., R. Mousavi, and C. Saydam, Predicting emergency medical service call demand: A modern spatiotemporal machine learning approach. Operations Research for Health Care, 2021. 28: p. 100285. 11. Neira-Rodado, D., et al., A novel machine learning approach for spatiotemporal prediction of EMS events: A case study from Barranquilla, Colombia. Heliyon, 2025. 11(2): p. e41904. 12. Channouf, N., et al., The application of forecasting techniques to modeling emergency medical system calls in Calgary, Alberta. Health Care Management Science, 2007. 10(1): p. 25–45. 13. Setzler, H., C. Saydam, and S. Park, EMS call volume predictions: A comparative study. Computers & Operations Research, 2009. 36(6): p. 1843–1851. 14. Hyndman, R.J. and A.B. Koehler, Another look at measures of forecast accuracy. International Journal of Forecasting, 2006. 22(4): p. 679–688. 15. Lee, H. and T. Lee, Demand modelling for emergency medical service system with multiple casualties cases: k-inflated mixture regression model. Flexible Services and Manufacturing Journal, 2021. 33(4): p. 1090–1115. 16. Matteson, D.S., et al., Forecasting emergency medical service call arrival rates. 2011. 17. Cressie, N. and C.K. Wikle, Statistics for spatio-temporal data. 2011: John Wiley & Sons. 18. Zhou, Z., et al., A spatio-temporal point process model for ambulance demand. Journal of the American Statistical Association, 2015. 110(509): p. 6–15. 19. Zhou, Z. and D.S. Matteson, Predicting Ambulance Demand: a Spatio-Temporal Kernel Approach, in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2015, Association for Computing Machinery: Sydney, NSW, Australia. p. 2297–2303. 20. Abreu, P., D. Santos, and A. Barbosa-Povoa, Data-driven forecasting for operational planning of emergency medical services. Socio-Economic Planning Sciences, 2023. 86: p. 101492. 21. Monks, T., et al., Forecasting the daily demand for emergency medical ambulances in England and Wales: a benchmark model and external validation. BMC Medical Informatics and Decision Making, 2023. 23(1): p. 117. 22. Tluli, R., et al., A Survey of Machine Learning Innovations in Ambulance Services: Allocation, Routing, and Demand Estimation. IEEE Open Journal of Intelligent Transportation Systems, 2024. 5: p. 842–872. 23. Shahidian, N., et al., Short-term forecasting of emergency medical services demand exploring machine learning. Computers & Industrial Engineering, 2025. 200: p. 110765. 24. Jin, R., et al., Predicting Emergency Medical Service Demand With Bipartite Graph Convolutional Networks. IEEE Access, 2021. 9: p. 9903–9915. 25. Megouo, T.G.P. and S. Pierre, A Stacking Ensemble Machine Learning Model for Emergency Call Forecasting. IEEE Access, 2024. 12: p. 115820–115837. 26. Garg, T., D. Toshniwal, and M. Parida, A meta-learning ensemble framework for robust and interpretable prediction of emergency medical services demand. Scientific Reports, 2025. 16(1): p. 2132. 27. Hermansen, A.H. and O.J. Mengshoel. Forecasting Ambulance Demand using Machine Learning: A Case Study from Oslo, Norway. in 2021 IEEE Symposium Series on Computational Intelligence (SSCI). 2021. 28. Fotheringham, A.S. and D.W.S. Wong, The Modifiable Areal Unit Problem in Multivariate Statistical Analysis. Environment and Planning A: Economy and Space, 1991. 23(7): p. 1025–1044. 29. Shi, X., et al., Convolutional LSTM network: A machine learning approach for precipitation nowcasting. Advances in neural information processing systems, 2015. 28. 30. Li, Y., et al., Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. arXiv preprint arXiv:1707.01926, 2017. 31. Wu, Z., et al., Graph wavenet for deep spatial-temporal graph modeling. arXiv preprint arXiv:1906.00121, 2019. 32. Zhang, C., et al., Deepmeshcity: A deep learning model for urban grid prediction. ACM Transactions on Knowledge Discovery from Data, 2024. 18(6): p. 1–26. 33. Kipf, T.N. and M. Welling, Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907, 2016. 34. Hamilton, W., Z. Ying, and J. Leskovec, Inductive representation learning on large graphs. Advances in neural information processing systems, 2017. 30. 35. Gneiting, T. and A.E. Raftery, Strictly Proper Scoring Rules, Prediction, and Estimation. Journal of the American Statistical Association, 2007. 102(477): p. 359–378. 36. Salinas, D., et al., DeepAR: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting, 2020. 36(3): p. 1181–1191. 37. Lim, B., et al., Temporal Fusion Transformers for interpretable multi-horizon time series forecasting. International Journal of Forecasting, 2021. 37(4): p. 1748–1764. 38. Saito, T. and M. Rehmsmeier, The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets. PLOS ONE, 2015. 10(3): p. e0118432. 39. Brodersen, K.H., et al. The Balanced Accuracy and Its Posterior Distribution. in 2010 20th International Conference on Pattern Recognition. 2010. 40. He, H. and E.A. Garcia, Learning from Imbalanced Data. IEEE Transactions on Knowledge and Data Engineering, 2009. 21(9): p. 1263–1284. 41. Davis, J. and M. Goadrich. The relationship between Precision-Recall and ROC curves. in Proceedings of the 23rd international conference on Machine learning. 2006. 42. Glenn, W.B., Verification of forecasts expressed in terms of probability. Monthly weather review, 1950. 78(1): p. 1–3. 43. Lambert, D., Zero-inflated Poisson regression, with an application to defects in manufacturing. Technometrics, 1992. 34(1): p. 1–14. 44. Chawla, N.V., et al., SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 2002. 16: p. 321–357. 45. Niculescu-Mizil, A. and R. Caruana, Predicting good probabilities with supervised learning, in Proceedings of the 22nd international conference on Machine learning. 2005, Association for Computing Machinery: Bonn, Germany. p. 625–632. 46. Guo, C., et al., On Calibration of Modern Neural Networks, in Proceedings of the 34th International Conference on Machine Learning, P. Doina and T. Yee Whye, Editors. 2017, PMLR: Proceedings of Machine Learning Research. p. 1321––1330. 47. Van Calster, B., et al., Calibration: the Achilles heel of predictive analytics. BMC Medicine, 2019. 17(1): p. 230. 48. Murphy, A.H., A New Vector Partition of the Probability Score. Journal of Applied Meteorology and Climatology, 1973. 12(4): p. 595–600. 49. Koenker, R. and G. Bassett, Regression Quantiles. Econometrica, 1978. 46(1): p. 33–50. 50. Khosravi, A., et al., Comprehensive Review of Neural Network-Based Prediction Intervals and New Advances. IEEE Transactions on Neural Networks, 2011. 22(9): p. 1341–1356. 51. Hochreiter, S. and J. Schmidhuber, Long Short-Term Memory. Neural Computation, 1997. 9(8): p. 1735–1780. 52. Vaswani, A., et al., Attention is all you need. Advances in neural information processing systems, 2017. 30. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/102206 | - |
| dc.description.abstract | 緊急醫療服務（Emergency Medical Services, EMS）需求具高度隨機性與時空異質性，若能準確掌握需求於不同時間與空間單元的變化，將有助於救護資源之前置部署與動態調度。本研究以臺北市為研究區，建構「時空網格化 × 機率式多步預測」之 EMS 需求預測框架，目的在於高解析度情境下同時提供點預測與不確定性資訊，以支援風險導向之資源規劃。
本研究以規則網格（1,000m × 1,000m）為空間單元，並以 4 小時為一時段進行時間切分，彙整各網格之需求量形成多序列時間序列資料；整合時間日曆特徵、氣象（降雨、氣溫）與人口結構（總人口、高齡人口）等共變數，其中降雨以 Kriging 內插至網格尺度，人口則由行政區轉換至網格以維持尺度一致。資料期間選取民國 107、108、112、113 年以降低非典型事件干擾。 模型採用 DeepAR 與 Temporal Fusion Transformer（TFT）進行機率預測，並以 MAE、RMSE、預測區間覆蓋率（PICP@80%）衡量準確性與覆蓋表現；另納入容忍誤差率（TRE，τ=±1）評估可操作性，並以分位數校準曲線檢核分位數輸出之校準程度。本研究貢獻在於建立高解析度網格需求之機率預測流程與多源資料整合方法，並比較 DeepAR 與 TFT 之系統性權衡，提供救護部署與決策支援之實證基礎。 | zh_TW |
| dc.description.abstract | Emergency Medical Services (EMS) demand is highly stochastic and spatiotemporally heterogeneous. Accurate forecasting can support ambulance pre-positioning and dynamic dispatching. This study proposes a grid-based probabilistic multi-horizon forecasting framework for EMS demand in Taipei City, providing both point forecasts and uncertainty information for risk-aware planning.
Taipei is partitioned into regular 1,000 m × 1,000 m grids, with demand aggregated into 4-hour intervals to form multiple time series. Covariates include calendar/time features, meteorological variables (rainfall and temperature), and demographics (total and older population). Station rainfall is interpolated to grids using Kriging, and administrative-area population is converted to grids to ensure spatial consistency. Data from 2018, 2019, 2023, and 2024 (ROC years 107, 108, 112, 113) are used to reduce atypical effects. We implement DeepAR and the Temporal Fusion Transformer (TFT) for probabilistic multi-horizon forecasting. Performance is evaluated by MAE, RMSE, and the 80% prediction interval coverage probability (PICP@80%), together with the tolerant rate error (TRE, τ=±1) to reflect operational usability. We further assess probabilistic reliability using quantile calibration curves. The results provide a reproducible high-resolution forecasting pipeline and an empirical comparison of DeepAR and TFT to inform EMS deployment and decision support. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2026-04-08T16:17:26Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2026-04-08T16:17:26Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 誌謝 II
摘要 III Abstract IV 目次 V 圖次 VII 表次 IX 1 第一章 簡介 1 1.1 研究背景與動機 2 1.2 研究問題 2 1.3 研究目的與研究範圍 3 1.4 研究方法概述 3 1.5 論文架構 4 2 第二章 文獻回顧 5 2.1 EMS需求預測經典與作業決策脈絡 5 2.2 現代ML／DL的時空EMS預測 7 2.3 機率預測與評估方法 9 2.4 本研究之方法基礎模型與研究定位 12 2.5 文獻評述與研究缺口 14 3 第三章 研究方法 18 3.1 研究架構 18 3.2 研究資料與研究範圍 19 3.3 資料前處理與特徵工程 21 3.4 模型輸入資料產製 28 3.5 預測模型與訓練設定（Modeling & Training） 32 3.6 模型評估指標（Evaluation Metrics） 33 3.7 置換法特徵重要性（Permutation Feature Importance） 35 4 第四章 實驗結果與討論 39 4.1 實驗設計總覽與評估框架 39 4.2 整體效能綜合比較（Overall Performance Comparison） 40 4.3 點預測準確性深度分析（In-depth Analysis of Point Prediction Accuracy） 41 4.4 機率式預測可靠性評估（Evaluation of Probabilistic Forecast Reliability） 44 4.5 實務應用效益分析：容忍誤差率（TRE） 49 4.6 置換法特徵重要性（Permutation Feature Importance） 53 4.7 綜合討論與模型特性總結 58 5 第五章 結論與未來研究方向 60 5.1 研究結論 60 5.2 研究貢獻 61 5.3 研究限制 62 5.4 未來研究方向 63 參考文獻 65 | - |
| dc.language.iso | zh_TW | - |
| dc.subject | EMS | - |
| dc.subject | 需求預測 | - |
| dc.subject | 時空網格 | - |
| dc.subject | 機率式預測 | - |
| dc.subject | DeepAR | - |
| dc.subject | TFT | - |
| dc.subject | EMS | - |
| dc.subject | demand forecasting | - |
| dc.subject | spatiotemporal grid | - |
| dc.subject | probabilistic forecasting | - |
| dc.subject | DeepAR | - |
| dc.subject | TFT | - |
| dc.title | 基於深度學習之臺北市緊急醫療服務需求之時空網格化機率預測：以DeepAR與TFT為例 | zh_TW |
| dc.title | Probabilistic Grid-Based EMS Demand Forecasting in Taipei: DeepAR and TFT | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 114-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 汪立本;林祺皓 | zh_TW |
| dc.contributor.oralexamcommittee | Li-Pen Wang;Chi-Hao Lin | en |
| dc.subject.keyword | EMS,需求預測時空網格機率式預測DeepARTFT | zh_TW |
| dc.subject.keyword | EMS,demand forecastingspatiotemporal gridprobabilistic forecastingDeepARTFT | en |
| dc.relation.page | 69 | - |
| dc.identifier.doi | 10.6342/NTU202600877 | - |
| dc.rights.note | 同意授權(全球公開) | - |
| dc.date.accepted | 2026-03-23 | - |
| dc.contributor.author-college | 工學院 | - |
| dc.contributor.author-dept | 土木工程學系 | - |
| dc.date.embargo-lift | 2026-04-09 | - |
| 顯示於系所單位： | 土木工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-114-2.pdf | 5.9 MB | Adobe PDF | 檢視/開啟 |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。