From forecast accuracy to inventory performance: Assessing machine-learning models for intermittent export demand

Asmuliardi Muluk; Taufik Nazir; Stephany Christie Lase; Dika Aulia Ardi

doi:10.65112/tcmis.100103

Authors

Asmuliardi Muluk Department of Industrial Engineering, Faculty of Engineering, Universitas Andalas, Padang, Indonesia https://orcid.org/0009-0008-9564-4867
Taufik Nazir Department of Industrial Engineering, Faculty of Engineering, Universitas Andalas, Padang, Indonesia https://orcid.org/0009-0008-9138-7808
Stephany Christie Lase Department of Industrial Engineering, Faculty of Engineering, Universitas Andalas, Padang, Indonesia https://orcid.org/0009-0008-7143-999X
Dika Aulia Ardi PT Digdaya Solusi, Padang, Indonesia https://orcid.org/0009-0001-8990-5716

DOI:

https://doi.org/10.65112/tcmis.100103

Keywords:

Intermittent demand, Demand forecasting, Inventory simulation, Periodic review policy, Machine learning, Forecast value

Abstract

Machine-learning forecasts are usually judged by accuracy, yet procurement decisions consume forecasts through an inventory policy whose lead times, review cycles, and lot sizes can overturn accuracy rankings. This paper quantified that value in replenishment decisions for an Indonesian cassia exporter's intermittent export demand. Ten approaches — the trading partner's company-described ex-ante annual plan, intermittent-demand estimators, ARIMA, and three machine-learning configurations — were embedded in a sequential simulation of the company's periodic-review policy over 55 months of grade-level demand, with a three-month lead time, lot-size constraints, and safety stocks calibrated to each method's realised errors. Each method was charged its cheapest configuration meeting a common achieved-fill requirement. Across sixteen demand units the plan and a Tweedie gradient-boosting model were not separated: the latter cost between 10% more and 4% less across four requirements, and the other eight alternatives at least 9% more, narrowing to 1% under a common warm-up. Resampling at the 0.98 requirement separated only two alternatives from the plan. Because the plan can carry the buyer's purchasing intentions, this shows no added value of history-based machine learning over such a plan in this portfolio, not a general absence of value. Separating the buffer rule from the forecast mattered more than the choice of model: replacing the fixed percentage mark-up with an error-calibrated buffer raised aggregate fill from 0.84 to 0.99 at unchanged cost, and from 0.75 to 0.99 at a fifth less. Policy simulation at a common service requirement should be the model-selection gate for intermittent-demand inventory systems.

Downloads

Download data is not yet available.

References

[1] S. A. Rajagukguk, E. Heriyanni, A. F. Maulida, and E. Putri, "Cross-modal attention transformer for multi-SKU demand forecasting in retail environments," in Information Systems for Intelligent Systems (Lecture Notes in Networks and Systems). Springer, 2026, pp. 281–290, doi: 10.1007/978-3-032-12993-2_27.

[2] A. Seyam, S. S. Mathew, M. E. Barachi, C. Zhang, and J. Shen, "The application of machine learning and deep learning on demand forecasting across time-critical industries: A systematic review," Adv. Eng. Inform., vol. 74, Sep. 2026, Art. no. 104625, doi: 10.1016/j.aei.2026.104625.

[3] I. Jackson, J. Namdar, M. J. Saénz, R. A. Elmquist III, and L. R. Dávila Novoa, "Revolutionize cold chain: An AI/ML driven approach to overcome capacity shortages," Int. J. Prod. Res., vol. 63, pp. 2190–2212, 2025, doi: 10.1080/00207543.2024.2398583.

[4] P. Eichenseer, L. Hans, and H. Winkler, "A data-driven machine learning model for forecasting delivery positions in logistics for workforce planning," Supply Chain Analytics, vol. 9, Mar. 2025, Art. no. 100099, doi: 10.1016/j.sca.2024.100099.

[5] T. U. Aye, A. A. Neto, and E. R. da Silva, "Knowing your surroundings: Tactical retail demand forecasting through exogeneity-aware machine learning," IEEE Eng. Manag. Rev., early access, 2025, doi: 10.1109/emr.2025.3645405.

[6] R. I. A. Omer, R. Hassan, and M. S. A. Aziz, "A review and comparative analysis of predictive models for supply chain demand forecasting," in Proc. 10th Int. Conf. Inf. Commun. Technol. Muslim World (ICT4M), Nov. 2025, pp. 1–7, doi: 10.1109/ict4m68001.2025.11363509.

[7] N. Messaoudi, and H. Fouraiji, "Improving supply chain performance by improving the quality of demand forecasting: Theoretical background and bibliometric analysis," in Proc. 16th Int. Conf. Logistics Supply Chain Manag. (LOGISTIQUA), May 2025, pp. 1–8, doi: 10.1109/logistiqua66323.2025.11122780.

[8] M. A. Habib, M. J. Hossain, S. Sakib, and M. M. Alam, "Bridging prediction and decision: An application-driven review of load forecasting in energy systems," Comput. Electr. Eng., vol. 134, Jun. 2026, Art. no. 111113, doi: 10.1016/j.compeleceng.2026.111113.

[9] M. Z. Babai, A. Syntetos, and R. Teunter, "Intermittent demand forecasting: An empirical study on accuracy and the risk of obsolescence," Int. J. Prod. Econ., vol. 157, pp. 212–219, Nov. 2014, doi: 10.1016/j.ijpe.2014.08.019.

[10] A. A. Syntetos, K. Nikolopoulos, and J. E. Boylan, "Judging the judges through accuracy-implication metrics: The case of inventory forecasting," Int. J. Forecast., vol. 26, no. 1, pp. 134–143, Jan. 2010, doi: 10.1016/j.ijforecast.2009.05.016.

[11] F. Petropoulos, X. Wang, and S. M. Disney, "The inventory performance of forecasting methods: Evidence from the M3 competition data," Int. J. Forecast., vol. 35, no. 1, pp. 251–265, Jan. 2019, doi: 10.1016/j.ijforecast.2018.01.004.

[12] N. Kourentzes, J. R. Trapero, and D. K. Barrow, "Optimising forecasting models for inventory planning," Int. J. Prod. Econ., vol. 225, Jul. 2020, Art. no. 107597, doi: 10.1016/j.ijpe.2019.107597.

[13] N. Sanders and G. Graman, "Quantifying costs of forecast errors: A case study of the warehouse environment," Omega, vol. 37, no. 1, pp. 116–125, Feb. 2009, doi: 10.1016/j.omega.2006.10.004.

[14] R. J. Hyndman and A. B. Koehler, "Another look at measures of forecast accuracy," Int. J. Forecast., vol. 22, no. 4, pp. 679–688, Oct. 2006, doi: 10.1016/j.ijforecast.2006.03.001.

[15] S. Kolassa, "Evaluating predictive count data distributions in retail sales forecasting," Int. J. Forecast., vol. 32, no. 3, pp. 788–803, Jul. 2016, doi: 10.1016/j.ijforecast.2015.12.004.

[16] J. D. Croston, "Forecasting and stock control for intermittent demands," J. Oper. Res. Soc., vol. 23, no. 3, pp. 289–303, Sep. 1972, doi: 10.1057/jors.1972.50.

[17] A. A. Syntetos and J. E. Boylan, "The accuracy of intermittent demand estimates," Int. J. Forecast., vol. 21, no. 2, pp. 303–314, Apr. 2005, doi: 10.1016/j.ijforecast.2004.10.001.

[18] R. H. Teunter, A. A. Syntetos, and M. Z. Babai, "Intermittent demand: Linking forecasting to inventory obsolescence," Eur. J. Oper. Res., vol. 214, no. 3, pp. 606–615, Nov. 2011, doi: 10.1016/j.ejor.2011.05.018.

[19] J. Boylan and A. Syntetos, Intermittent Demand Forecasting: Context, Methods and Applications. Hoboken, NJ, USA: Wiley, 2021, doi: 10.1002/9781119135289.

[20] S. Makridakis, E. Spiliotis, and V. Assimakopoulos, "M5 accuracy competition: Results, findings, and conclusions," Int. J. Forecast., vol. 38, no. 4, pp. 1346–1364, Oct. 2022, doi: 10.1016/j.ijforecast.2021.11.013.

[21] Y. Hu et al., "Implementing a prediction driven framework for emergency department nurse staffing to optimize real time decisions," npj Health Syst., vol. 2, May 2025, Art. no. 16, doi: 10.1038/s44401-025-00019-2.

[22] M. F. Hadi and A. Muluk, "A machine-learning decision support system for demand-driven nurse staffing in outpatient clinics: Design and validation at an Indonesian public hospital," Trans. Comput. Model. Intell. Syst., vol. 4, Jul. 2026, Art. no. 10092, doi: 10.65112/tcmis.10092.

[23] S. Maitra, "Agentic automation driven reinforcement learning for inventory optimization," J. Model. Manag., early access, pp. 1–36, Apr. 2026, doi: 10.1108/jm2-07-2025-0347.

[24] K. Ramaekers and G. K. Janssens, "Optimal policies for demand forecasting and inventory management of goods with intermittent demand," J. Appl. Oper. Res., vol. 6, no. 2, pp. 111–123, 2014.

[25] A. A. Syntetos, J. E. Boylan, and J. D. Croston, "On the categorization of demand patterns," J. Oper. Res. Soc., vol. 56, no. 5, pp. 495–503, May 2005, doi: 10.1057/palgrave.jors.2601841.

[26] N. Kourentzes, "On intermittent demand model optimisation and selection," Int. J. Prod. Econ., vol. 156, pp. 180–190, Oct. 2014, doi: 10.1016/j.ijpe.2014.06.007.

[27] S. C. Lase, "Pengembangan sistem pendukung pengadaan bahan baku berbasis peramalan permintaan menggunakan machine learning (studi kasus: PT XYZ) [Development of a raw-material procurement support system based on machine-learning demand forecasting (case study: PT XYZ)]," Undergraduate thesis, Dept. Ind. Eng., Univ. Andalas, Padang, Indonesia, 2026.