A Comparative Analysis of Clustering Schemes and Optimal Forecasting Models for Spare Parts Demand Forecast in Maintenance Operation
Main Article Content
Abstract
This study aims to identify the most appropriate clustering scheme and forecasting method for spare parts used in maintenance operations with heterogeneous demand characteristics. The proposed methodology first classifies spare parts based on the Average Demand Interval (ADI) and the Squared Coefficient of Variation (CV²). Four clustering techniques are compared: demand segmentation based on demand patterns, K-means clustering, hierarchical clustering, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The empirical analysis is conducted using monthly demand data from 2020 to 2023 (48 months) covering 1,735 spare parts. Demand data from 2020 to 2022 (36 months) are used as the training set for model development and parameter calibration, while data from 2023 (12 months) are used as the test set to evaluate out-of-sample forecasting performance. For each cluster, demand is forecast using the Simple Moving Average (SMA), Holt’s exponential smoothing method, Croston’s method, and the Autoregressive Integrated Moving Average (ARIMA) model. Forecast accuracy is evaluated using the Symmetric Mean Absolute Percentage Error (sMAPE). The results indicate that DBSCAN combined with the ARIMA model provides the lowest sMAPE values across all clusters. Specifically, cluster 1, cluster 2, cluster 3, and cluster 4 achieve sMAPE values of 3.01%, 27.03%, 5.37%, and 4.17%, respectively. These findings demonstrate the effectiveness of integrating clustering and forecasting techniques for spare-parts demand forecasting in maintenance operations.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The articles published are the opinion of the author only. The author is responsible for any legal consequences. That may arise from that article.
References
Krungsri Research. (2021). Industry Outlook 2021–2023,” Krungsri.com. [Online] Available: https://www.krungsri.com/th/research/industry/summary-outlook/industry-summary-outlo ok-2021-2023 (in Thai).
M. Seyedan, F. Mafakheri, and C. Wang, “Cluster-based demand forecasting using Bayesian model averaging: An ensemble learning approach,” Decision Analytics Journal, vol. 3, 2022, Art. no. 100033, doi: 10.1016/j.dajour.2022.100033.
T. Chunpong, “Inventory management for integrated circuit packaging materials under variable demand,” Master’s thesis, Department of Industrial Engineering Faculty of Engineering Chulalongkorn University, Bangkok, Thailand, 2021 (in Thai).
S. Wongkamphu, “Battery sales forecasting using traditional forecasting and machine learning techniques,” Master’s thesis, Faculty of Engineering Chulalongkorn University, Bangkok, Thailand, 2023 (in Thai).
N. Paranthaman, H. N. Perera, N. T. Thalagala, and D. Kosgoda, “Evaluating intermittent demand forecasting techniques for spare part supply chains,” IFAC-PapersOnLine, vol. 58, no. 14, pp. 136–141, 2024, doi: 10.1016/j.ifacol.2024.09.025.
T. B. Affonso, S. V. Conceição, L. R. Muniz, J. F. D. F. Almeida, and J. C. Lima, “A new hybrid forecasting method for spare part inventory management using heuristics and bootstrapping,” Decision Analytics Journal, vol. 10, 2024, Art. no. 100415, doi: 10.1016/j.dajour.2024.100415.
M. İfraz, A. Aktepe, S. Ersöz, and T. Çetinyokuş, “Demand forecasting of spare parts with regression and machine learning methods: Application in a bus fleet,” Journal of Engineering Research, vol. 11, no. 2, 2023, Art. no. 100057, doi: 10.1016 /j.jer.2023.100057.
A. A. Syntetos, J. E. Boylan, and J. D. Croston, “On the categorization of demand patterns,” Journal of the Operational Research Society, vol. 56, no. 5, pp. 495–503, 2005, doi: 10.1057/palgrave.jors.2601841.
B. A. Kannan, G. Kodi, O. Padilla, D. Gray, and B. C. Smith, “Forecasting spare parts sporadic demand using traditional methods and machine learning – a comparative study,” SMU Data Science Review, vol. 3, no. 2, 2020, Art. no. 9.
D. Visbal-Cadavid, E. Delahoz-Domínguez, and A. Mendoza-Mendoza, “A multiple factor analysis and hierarchical clustering of global logistics governance and development,” Decision Analytics Journal, vol. 15, 2025, Art. no. 100579, doi: 10.1016/ j.dajour.2025.100579.
R. Ma, J. Sha, S. Zhang, D. Zhu, W. Kang, and J. Liu, “Fast grouping fusion method of dual carbon monitoring data based on DBSCAN clustering algorithm,” Results in Engineering, vol. 26, 2025, Art. no. 105057, doi: 10.1016/j.rineng.2025.105057.
A. Oyemaja, “Moving averages in time series analysis: Understanding trends & forecasting,” ResearchGate Preprint, Mar. 2024, doi: 10.13140/RG.2.2.14811.17441.
W. Dechadumrongchai, “Improvement of demand forecasting and replenishment policy for agricultural machine spare parts,” Master’s thesis, Faculty of Engineering Chulalongkorn University, Bangkok, Thailand, 2021 (in Thai).
M. Doszyn, “Accuracy of intermittent demand forecasting systems in the enterprise,” European Research Studies Journal, vol. 23, no. 4, pp. 912–930, 2020, doi: 10.35808/ersj/1723.
T. Wongchitsukkasem, “Thailand steel scrap price forecasting using decomposition- based hybrid model,” Master’s thesis, Department of Industrial Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand, 2023 (in Thai).
S. Makridakis, E. Spiliotis, and V. Assimakopoulos, “The M4 competition: 100,000 time series and 61 forecasting methods,” International Journal of Forecasting, vol. 36, no. 1, pp. 54–74, 2020, doi: 10.1016/j.ijforecast.2019.04
