準確的需求預測對於數據驅動的供應鏈管理至關重要，在供應鏈管
理中，規劃人員必須將快速變化的大量數據轉化為可操作的見解，以應對
波動。然而，依賴定義不清的 "黑盒 "模型有損預測的可靠性。這項研究表
明，與傳統的統計方法相比，先進的機器學習技術可以顯著提高預測準確
性。利用台灣的菸草銷售數據，XGBoost（一種複雜的梯度提升機器學習
模型）被應用於需求預測。與基準季節指數平滑模型和 SARIMA 模型相比，
XGBoost 由於能靈活捕捉複雜的多變量模式，其均值平方根誤差和均值絕
對誤差降低了 38% 以上。研究結果證實了透明、適應性強的機器學習方法
優於僵化的統計技術。對於規劃複雜的現實世界需求而言，機器學習比傳
統的建模方法具有更高的精確度，透過現代人工智慧預測技術為增強資料
驅動型決策提供了途徑。
關鍵詞: 需求預測, 季節性指數平滑模型,SARIMA 模型, 機器學習模
型,XGBoost。

Accurate demand forecasting is critical for data-driven supply chain
management, where planners must translate massive, rapidly changing data into
actionable insights to tackle volatility. However, reliance on ill-defined "black
box" models undermines predictive reliability. This research demonstrates that
advanced machine learning techniques can significantly improve forecast accuracy
over traditional statistical approaches. Using tobacco sales data from Taiwan,
XGBoost, a sophisticated gradient boosting machine learning model, is applied for
demand forecasting. Compared to baseline Seasonal Exponential Smoothing and
SARIMA models, XGBoost achieves over 38% lower Root Mean Squared Error
and Mean Absolute Error due to its flexibility in capturing complex multivariate
patterns. The study results confirm the superiority of transparent, adaptable
machine learning methods over rigid statistical techniques. For planning complex
real-world demand, machine learning delivers substantially higher precision than
dated modeling traditions, providing a path to enhanced data-driven decision
making through modern AI-powered forecasting.
Keywords: Demand Forecasting, Seasonal Exponential Smoothing model,
SARIMA model, Machine Learning model, XGBoost.

Contents
ABSTRACT iii
摘要 iv
ACKNOWLEDGMENTS v
LIST OF TABLES viii
LIST OF FIGURES ix
1. INTRODUCTION 1
1.1 Background 1
1.2 Motivations 2
1.3 Research Approach 3
1.4 Research Objectives 5
2. LITERATURE REVIEW 9
2.1 Demand Forecasting in Supply Chain Management 9
2.1.1 The Bullwhip Effect 11
2.2 Demand Forecasting Techniques 14
2.2.1 Traditional Methods 16
2.2.2 Machine Learning Methods 16
2.2.3 Forecast Accuracy Measurements 18
2.3 Current Gaps and Challenges 19
3. METHODOLOGY 21
3.1 Determining the Research Questions and Objectives 22
3.2 Data Collection 23
3.3 Data Cleaning and Preprocessing 25
3.4 Data Analysis 26
3.4.1 Seasonal Exponential Smoothing Model 28
3.4.2 SARIMA Model 30
3.4.3 XGBoost Model 31
3.5 Interpretation of Results 34
4. DATA AND RESULTS 35
4.1 Data 35
4.2 Results 43
4.2.1 Seasonal Exponential Smoothing Model 43
4.2.2 SARIMA Model 45
4.2.3 XGBoost Model 46
4.3 Comparative Analysis 49
5. DISCUSSION AND CONCLUSION 52
5.1 Discussion 52
5.2 Conclusion 53
References 56
Appendix I: Original Data 62

References
Tai, Y. M., Ho, C. F., & Wu, W. H. (2010). The performance impact of implementing web-based eprocurement systems. International Journal of Production Research, 48(18), 5397-5414.
Wang, G., Gunasekaran, A., Ngai, E. W., & Papadopoulos, T. (2016). Big data analytics in logistics and supply chain management: Certain investigations for research and applications. International Journal of Production Economics, 176, 98-110.
Liu, Q., Sun, S. X., Wang, H., & Zhao, J. (2011). A multi-agent-based system for e-procurement exception management. Knowledge-Based Systems, 24(1), 49–57.
Giunipero, L. C., & Aly Eltantawy, R. (2004). Securing the Upstream Supply Chain: A Risk Management Approach. International Journal of Physical Distribution & Logistics Management, 34(9), 698-713. doi: 10.1108/09600030410567478.
Chen, F., Drezner, Z., Ryan, J. K., & Simchi-Levi, D. (2000). Quantifying the Bullwhip Effect in a Simple Supply Chain: The Impact of Forecasting, Lead Times, and Information. Management science, 46(3), 436-443.
Breitschwerdt, D., Cornet, A., Kempf, S., Michor, L., & Schmidt, M. (2017). The changing aftermarket game–and how automotive suppliers can benefit from arising opportunities. Advanced Industries, June. Retrieved from https://clepa.eu/wpcontent/uploads/2017/07/McKinsey_The-changing-aftermarket-game.pdf.
Hu, Q., Boylan, J. E., Chen, H., & Labib, A. (2018). OR in spare parts management: A review. European Journal of Operational Research, 266(2), 395–414.
Syntetos, A. A., Babai, M. Z., Boylan, J. E., Kolassa, S., & Nikolopoulos, K. (2016). Supply chain forecasting: Theory, practice, their gap and the future. European Journal of Operational Research, 252(1), 1–26.
Hiscock, R. & M. J. Bloomfield (2021). "The value of studying supply chains for tobacco control." Tob Prev Cessat 7: 15.
Hauke, J., Lorscheid, I., & Meyer, M. (2018). Individuals and their interactions in demand-planning processes: An agent-based, computational testbed. International Journal of Production Research, 56(13), 4644–4658.

Zhou, H., W. Benton, D. A. Schilling & G. W. Milligan. (2011). Supply Chain Integration and the SCOR Model. Journal of Business Logistics 32 (4): 332–344.
Wang, X., & Petropoulos, F. (2016). To select or to combine? The inventory performance of model and expert forecasts. International Journal of Production Research, 54(17), 5271–5282.
Chopra, S., & Meindl, P. (2010). Supply chain management: strategy, planning, and operation. Boston: Prentice Hall, c2010.
Moon, M. A., Mentzer, J. T., & Smith, C. D. (2003). Conducting a sales forecasting audit. International Journal of Forecasting, 19(1), 5–25.
Halldórsson, Á., G. Stefánsson, P. Jonsson, L. Kjellsdotter, and M. Rudberg. (2007). “Applying Advanced Planning Systems for Supply Chain Planning: Three Case Studies.” International Journal of Physical Distribution & Logistics Management 37 (10): 816–834.
Oliva, R., and N. Watson. (2011). “Cross-functional Alignment in Supply Chain Planning: A Case Study of Sales and Operations Planning.” Journal of Operations Management 29 (5): 434–448.
Kaipia, R., J. Holmström, J. Småros, and R. Rajala. (2017). “Information Sharing for Sales and Operations Planning: Contextualized Solutions and Mechanisms.” Journal of Operations Management 52: 15–29.
Fildes, R., P. Goodwin, M. Lawrence, and K. Nikolopoulos. (2009). “Effective Forecasting and Judgmental Adjustments: An Empirical Evaluation and Strategies for Improvement in Supplychain Planning.” International Journal of Forecasting 25 (1): 3–23.
Zoryk-Schalla, A. J., J. C. Fransoo, and T. G. de Kok. (2004). “Modeling the Planning Process in Advanced Planning Systems.” Information & Management 42 (1): 75–87.
Moritz, B., E. Siemsen, and M. Kremer. (2014). Judgmental Forecasting: Cognitive Reflection and Decision Speed. Production and Operations Management 23 (7): 1146–1160.
Chybalski, F. (2017). Forecast value added (FVA) analysis as a means to improve the efficiency of a forecasting process. Operations Research and Decissions; ISSN 2081-8858.
Naish, H.F. (1994). Production smoothing in the linear quadratic inventory model. Quarterly Journal of Economics, 104, 864-875.
Terwiesch, C., Ren, T., Ho, H., & Cohen, M. (2005). Forecast sharing in the semiconductor equipment supply chain. Management Science, 51, 208-220.
McCarthy, T.M., Davis, D.F., Golicic, S.L., & Mentzer, J.T. (2006). The evolution of sales forecasting management: a 20-year longitudinal study of forecasting practices. Journal of Forecasting, 25 (5). 303-324.
Makridakis, S., Anderson, A., Carbone, R., Fildes, R., Hibon, M., Lewandowski, R., Newton, J., Parzen, E., & Winkler, R. (1982). The accuracy of extrapolation (time series) methods: results of a forecasting competition. Journal of Forecasting, 1, 111-153.
Makridakis, S., & Hibon, M. (2000). The M3-Competition: results, conclusions and implications. International journal of forecasting, 16(4), 451-476.
Chatfield, C. & Yar, M. (1988), Holt-Winters forecasting: some practical issues. Journal of the Royal Statistical Society, 37 (2), 129-140.
Chopra, S., & Meindl, P. (2016). Supply chain management: Strategy, planning, and operation. Pearson.
Siami-Namini, S., Dehkordi, H. T., & Hu, M. (2021). Seasonal forecasting: a review of algorithms, techniques, and applications. Annals of Operations Research, 298(1-2), 299-338.
Gardner, E. S. (2006). Exponential smoothing: The state of the art. Journal of Forecasting, 25(1), 1-28.
Bandara, D., Chiaraviglio, L., Granelli, F., & Rashwand, H. (2019). Seasonal time series forecasting with long short-term memory networks. IEEE Transactions on Neural Networks and Learning Systems, 30(12), 3861-3875.
Silver, E.A., Pyke, D.F., & Peterson, R. (1998). Inventory Management and Production Planning and Scheduling, 3rd ed., New York: John Wiley & Sons.
Lee, H.L., Padmanabhan, V., & Whang, S. (2004). Comments on information distortion in a supply chain: The bullwhip effect - The bullwhip effect: Reflections. Management Science, 50 (12), 1887-1893.
Towill, D.R., Zhou, L., & Disney, S.M. (2007). Reducing the bullwhip effect: looking through the appropriate lens. International Journal of Production Economics, 108, 444-453.
Lee, H. L., Padmanabhan, V., & Whang, S. (1997). Information distortion in a supply chain: The bullwhip effect. Management Science, 43 (4), 546-559.
Forrester, J.W. (1961). Industrial Dynamics. Productivity Press, Portland OR.
Sterman, J. D. (1989). Modeling managerial behavior: Misperceptions of feedback in a dynamic decision making experiment. Management Science, 35 (3). 321-339.
Burbidge, J.L. (1981). The new approach to production. Production Engineer, 40 (12), 769-784.
Eichenbaum, M.S. (1989). Some empirical evidence of the production level and production cost smoothing models of inventory investment. American Economics Review, 79 (4), 853-864.
Blackburn, J. D. (1991). The quick response movement in the apparel industry: A case study in time-compressing supply chains. In: Blackburn, J.D.(Ed.). Time Based Competition: The Next Battleground in American Manufacturing. Irwin, Homewood, IL (Chapter 11).
Caplice, C., & Sheffi, Y. (2006). Demand Forecasting I: Time Series Analysis. In ESD.260J Logistics Systems. Fall 2006. Massachusetts Institute of Technology: MIT OpenCourseWare.
Green, K. C., & Armstrong, J. S. (2012). Demand Forecasting: Evidence-Based Methods. Available at SSRN 3063308.
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and Practice (2nd ed.): OTexts.
Nerlove, M., & Diebold, F. X. (1990). Autoregressive and Moving-average Time-series Processes. In J. Eatwell, M. Milgate, & P. Newman (Eds.), Time Series and Statistics (pp. 25-35). London: Palgrave Macmillan UK.
Weller, M., & Crone, S. F. (2012). Supply Chain Forecasting: Best Practices & Benchmarking Study: Lancaster Centre for Forecasting.
Mitrea, C. A., Lee, C. K. M., & Wu, Z. (2009). A Comparison between Neural Networks and Traditional Forecasting Methods: A Case Study. International Journal of Engineering Business Management, 1, 11. doi:10.5772/6777.
Carbonneau, R., Laframboise, K., & Vahidov, R. (2008). Application of Machine Learning Techniques for Supply Chain Demand Forecasting. European Journal of Operational Research, 184(3), 1140-1154. doi:https://doi.org/10.1016/j.ejor.2006.12.004.
Crone, S. F., Hibon, M., & Nikolopoulos, K. (2011). Advances in Forecasting with Neural Networks? Empirical Evidence from the NN3 Competition on Time Series Prediction. International Journal of Forecasting, 27(3), 635-660.doi:https://doi.org/10.1016/j.ijforecast.2011.04.001
Tashman, Leonard J. (2000). Out-of-sample tests of forecasting accuracy: An analysis and review. International Journal of Forecasting 16: 437–50.
Kolassa, S. (2016). Evaluating predictive count data distributions in retail sales forecasting. International Journal of Forecasting, 32(3), 788–803. https://doi.org/10.1016/j.ijforecast.2015.12.004
Schwertman, N. C., Gilks, A. J., & Cameron, J. (1990). A simple noncalculus proof that the median minimizes the sum of the absolute deviations. The American Statistician, 44, 38–39. https://doi.org/10.1080/00031305.1990.10475690
Hyndman, R. J., & Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4), 679–688. https://doi.org/10.1016/j.ijforecast.2006.03.001
Armstrong, J. S. (2001). Standards and practices for forecasting. In Principles of forecasting (pp. 679–732). Springer.
Chen, C., Twycross, J., & Garibaldi, J. M. (2017). A new accuracy measure based on bounded relative error for time series forecasting. PloS One, 12(3), e0174202. https://doi.org/10.1371/journal.pone.0174202
Boone, T., Boylan, J. E., Fildes, R., Ganeshan, R., & Sanders, N. (2019). Perspectives on Supply Chain Forecasting. International Journal of Forecasting, 35(1), 121-127. doi:https://doi.org/10.1016/j.ijforecast.2018.11.002.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2018). Multivariate Data Analysis (8th ed.). Cengage Learning.
Leedy, P. D., Ormrod, J. E., & Singleton, M. (2019). Practical research: Planning and design. Pearson Higher Ed.
Creswell, J. W., & Creswell, J. D. (2017). Research design: Qualitative, quantitative, and mixed methods approaches. Sage publications.
Field, A. (2018). Discovering statistics using IBM SPSS statistics. SAGE Publications.
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An Introduction to Statistical Learning (Vol. 112, p. 18). New York: springer.
Kuhn, M., & Johnson, K. (2013). Applied predictive modeling (pp. 61-62). New York: Springer.