The application of machine learning algorithms in the estimation of production lead times: A case study of a steel

Abstract

In companies with a diverse range of products, it can be challenging to regulate production times. Having an understanding of production lead times is crucial for addressing issues such as deadlines, cost, production scheduling, and customer satisfaction. It is challenging for the company to provide the customer with an accurate estimation of the time required to produce and deliver a product that has not been produced before. One of the advantages of knowing the production times is to be able to adjust the machines to use them more efficiently when preparing the production plan. In this study, input data were obtained from a steel construction company, including dimensions such as size, diameter, and weight. Additionally, the times required for the production of different products were measured. Based on these times, production times were estimated using machine learning algorithms, including Decision Tree, Random Forest, and Gradient Boosting. Consequently, precise predictions were generated with an accuracy rate of approximately 96.9%. A test data set was then created with the objective of estimating the time required to produce a product that has never been produced. Additionally, the times of products that have not yet been produced were estimated. For each new product ordered, the machine must be adjusted and calibrated separately, which represents a significant loss of time and cost for the company. The objective of this research is to develop a model that can predict the time required to deliver a new product once it has been ordered. Furthermore, the aim is to enhance the efficiency of machine utilization.

Keywords

• Machine learning
• production time
• random forest regressor
• regression
• steel construction.

References

1. REFERENCES
2. [1] Işık, K., & Ulusoy, S. K. (2021). Determination of factors affecting production times in the metal sector by data mining methods. Journal of Gazi University Faculty of Engineering and Architecture, 36(4), 19491962. [CrossRef]
3. [2] Gökler, S. H., Ulus, İ., Cihat, F., & Boran, S. (2015). Estimation of machine part replacement times based on a cost model using two-parameter Weibull distribution. Beykent University Journal of Science and Engineering, 8(2), 6184. [CrossRef]
4. [3] Eker, R., Alkiş, K. C., Uçar, Z., & Aydın, A. (2023). The use of machine learning in forestry. Turkish Journal of Forestry, 24(2), 150177. [CrossRef]
5. [4] Sinap, V. (2024). Comparative performance analysis of machine learning algorithms in the retail sector: Black Friday sales forecasting. Selçuk University Social Sciences Vocational School Journal, 27(1), 6590. [CrossRef]
6. [5] Danacıoğlu, Ş. (2023). Evaluation of different machine learning algorithms for land cover mapping: The case of Izmir Province. Turkish Journal of Geography, (84), 105117. [CrossRef]
7. [6] Görentaş, M. B., & Uçkan, T. (2023). Clustering court decisions using machine learning methods. Computer Science, 8(2), 148158.
8. [7] Melekoğlu, E. (2023). Development of a new machine learning-based method for the diagnosis of chronic obstructive pulmonary disease (Doctoral dissertation). Available from National Thesis Center. (Accession No. 789510.)

9. [8] Lim, Z. H., Yusof, U. K., & Shamsudin, H. (2019). Manufacturing lead time classification using support vector machine. In: Badioze Zaman, H., et al. Advances in Visual Informatics: 6th International Visual Informatics Conference, IVIC 2019, Bangi, Malaysia, November 19–21, 2019, Proceedings 6 (pp. 268278). Springer International Publishing. [CrossRef]
10. [9] Schneckenreither, M., Haeussler, S., & Gerhold, C. (2021). Order release planning with predictive lead times: A machine learning approach. International Journal of Production Research, 59(11), 32853303. [CrossRef]
11. [10] Yüce, T., & Kabak, M. (2021). Predicting production times for detailed production areas at work center level using machine learning algorithms. Erciyes University Journal of the Institute of Science and Technology, 37(1), 4760.
12. [11] Haeussler, S., & Netzer, P. (2020). Comparison between rule-and optimization-based workload control concepts: A simulation optimization approach. International Journal of Production Research, 58(12), 37243743. [CrossRef]
13. [12] El Mekkaoui, S., Benabbou, L., & Berrado, A. (2022). Machine learning models for efficient port terminal operations: Case of vessels’ arrival times prediction. IFAC-PapersOnLine, 55(10), 31723177. [CrossRef]
14. [13] Agwu, O. E., Alkouh, A., Alatefi, S., Azim, R. A., & Ferhadi, R. (2024). Utilization of machine learning for the estimation of production rates in wells operated by electrical submersible pumps. Journal of Petroleum Exploration and Production Technology, (14), 1205–1233. [CrossRef]
15. [14] Alsakka, F., Yu, H., El-Chami, I., Hamzeh, F., & Al-Hussein, M. (2024). Digital twin for production estimation, scheduling and real-time monitoring in offsite construction. Computers & Industrial Engineering, 191, Article 110173. [CrossRef]
16. [15] Gyulai, D., Pfeiffer, A., Bergmann, J., & Gallina, V. (2018). Online lead time prediction supporting situation-aware production control. Procedia CIRP, 78, 190195. [CrossRef]
17. [16] Chen, S., Li, X., Liu, R., & Zeng, S. (2019). Extension data mining method for improving product manufacturing quality. Procedia Computer Science, 162, 146155. [CrossRef]
18. [17] Dehghani, M., Jahani, S., & Ranjbar, A. (2024). Comparing the performance of machine learning methods in estimating the shear wave transit time in one of the reservoirs in southwest of Iran. Scientific Reports, 14(1), Article 4744. [CrossRef]
19. [18] Schonlau, M., & Zou, R. Y. (2020). The random forest algorithm for statistical learning. The Stata Journal, 20(1), 329. [CrossRef]
20. [19] Das, O., & Bagci Das, D. (2022). Free vibration analysis of isotropic plates using regressive ensemble learning. European Journal of Science and Technology, (38), 428434.
21. [20] Çelik, S., & Özdemir, D. (2023). Bitcoin price prediction using random forest regression algorithm. Journal of Scientific Reports-B, (008), 5564.
22. [21] Rokach, L., & Maimon, O. (2014). Data mining with decision trees: Theory and applications (2nd ed.). World Scientific Publishing Co. Pte. Ltd. [CrossRef]
23. [22] Doğruel, M., & Ümit Fırat, S. (2021). Estimation of countries' innovation values using data mining decision trees and a comparative application with linear regression model. Istanbul Business Research, 50(2), 465493.
24. [23] Özkan, B., Parim, C., & Çene, E. (2023). Prediction of countries' development levels using decision tree and random forest methods. EKOIST Journal of Econometrics and Statistics, (38), 87104. [CrossRef]
25. [24] Çelik, S., Çetinkaya Bozkurt, Ö., & Ekşili, N. (2022). Determination of expressions in employee performance scale with decision tree algorithm. Journal of Mehmet Akif Ersoy University Economics and Administrative Sciences Faculty, 9(1), 561584. [CrossRef]
26. [25] Rao, H., Shi, X., Rodrigue, A. K., Feng, J., Xia, Y., Elhoseny, M., & Yuan, X. (2019). Feature selection based on artificial bee colony and gradient boosting decision tree. Applied Soft Computing Journal, 74, 634642. [CrossRef]
27. [26] Kelle, A. C., & Yüce, H. (2022). Classification of DoS attacks on MQTT traffic using machine learning and interpretation of the model with SHAP. Journal of Materials and Mechatronics: A, 3(1), 5062. [CrossRef]
28. [27] Yeşilyurt, S., & Dalkılıç, H. (2021). Daily river flow prediction with Xgboost and gradient boost machine. In 3rd International Symposium of III Engineering Applications on Civil Engineering and Earth Sciences, Karabük, Turkey.
29. [28] Chai, T., & Draxler, R. R. (2014). Root mean square error (RMSE) or mean absolute error (MAE)?. Geoscientific Model Development, 7, 12471250. [CrossRef]
30. [29] Özen, N. S., Saraç, S., & Koyuncu, M. (2021). Prediction of COVID-19 cases with machine learning algorithms: The case of the United States. European Journal of Science and Technology, (22), 134139.