A Predict-then-Optimize Job Assignment Framework for Efficient Trademark Application Evaluation

Öz

Timely evaluation of trademark applications is critical for effective intellectual property protection, yet many national offices face rising workloads and limited staffing flexibility. At TURKPATENT, these pressures have contributed to growing backlogs and high variability in evaluation times. This study proposes a data-driven job assignment framework that integrates machine learning based prediction of examiner-specific completion times with a rolling-horizon mixed-integer programming model. A comprehensive feature set incorporating text indicators, classification codes, examiner effects, and workload characteristics is used to train predictive models, and predicted durations are subsequently embedded into an assignment model designed to minimize tardiness while maintaining workload balance. Multiple assignment policies are evaluated through extensive simulation under varying workload and prediction-quality scenarios. The results show that the predictive rolling-horizon model reduces tardy jobs by roughly 49.5% relative to the current system. A fully automated operational pipeline was implemented to enable daily deployment, and a real-world pilot with eight examiners achieved a 35.3% reduction in tardy jobs. The study demonstrates that integrating predictive analytics with optimization can substantially improve performance in administrative workflows and offers a scalable approach for modernizing trademark examination processes.

Anahtar Kelimeler

• Predict-then-optimize
• Job Assignment Optimization
• Predictive Modeling
• Rolling Horizon Optimization
• Trademark Application

Veriye Dayalı Tahmin ve Optimizasyon Yaklaşımıyla Etkin Marka Başvurusu Değerlendirmesi

Öz

Marka başvurularının zamanında değerlendirilmesi, etkin fikri mülkiyet koruması için kritik öneme sahiptir; ancak birçok ulusal ofis artan iş yükleri ve sınırlı personel esnekliği nedeniyle önemli zorluklarla karşılaşmaktadır. TÜRKPATENT’te bu sıkıntılar, büyüyen iş birikmelerine ve değerlendirme sürelerinde yüksek değişkenliğe yol açmıştır. Bu çalışma, değerlendirme sürelerini makine öğrenimi ile tahmin eden bir modeli, bir karma tamsayılı programlama modeliyle bütünleştiren veri odaklı bir iş atama çerçevesi önermektedir. Metin göstergeleri, sınıflandırma kodları, değerlendirici etkileri ve iş yükü özelliklerini içeren kapsamlı bir öznitelik seti ile tahmin modelleri eğitilmiş; tahmin edilen süreler ise gecikmeleri azaltırken iş yükü dengesini koruyacak şekilde tasarlanmış atama modeline entegre edilmiştir. Farklı iş yükü ve tahmin doğruluğu senaryoları altında farklı atama politikaları simülasyon modeli aracılığı ile değerlendirilmiştir. Sonuçlar, önerilen tahminle-ardından-eniyile modelinin mevcut sisteme kıyasla geciken iş sayısını yaklaşık %49,5 oranında azalttığını göstermektedir. Sistemin günlük kullanımını mümkün kılmak için tam otomatik bir operasyonel altyapı uygulanmış ve sekiz incelemeciyle yapılan gerçek dünya pilot çalışmasında geciken işlerde %35,3’lük bir azalma elde edilmiştir. Bu çalışma, öngörüsel analitik ile eniyilemenin bütünleştirilmesinin idari iş akışlarında performansı önemli ölçüde artırabileceğini ve marka inceleme süreçlerini modernize etmek için ölçeklenebilir bir yaklaşım sunduğunu göstermektedir.

Anahtar Kelimeler

• veriye dayalı
• marka başvurusu
• iş atama optimizasyonu
• tahminle-ardından-eniyile

Kaynakça

1. Anurag. (2018). Random forest analysis in ML and when to use it. Retrieved from https://www.newgenapps.com/blogs/random-forest-analysis-in-ml-and-when-to-use-it-2/ Bertsimas, D., & Kallus, N. (2019). From Predictive to Prescriptive Analytics. Management Science, 66(3), 1025–1044. https://doi.org/10.1287/mnsc.2018.3253
2. Cattrysse, D. G., & Van Wassenhove, L. N. (1992). A survey of algorithms for the generalized assignment problem. European Journal of Operational Research, 60(3), 260–272. https://doi.org/10.1016/0377- 2217(92)90077-M
3. Elmachtoub, A. N., & Grigas, P. (2022). Smart “Predict, then Optimize”. Management Science, 68(1), 9–26. https://doi.org/10.1287/mnsc.2020.3922
4. Geng, H., Ruan, H., Wang, R., Li, Y., Wang, Y., Chen, L., & Yan, J. (2024). Benchmarking PtO and PnO Methods in the Predictive Combinatorial Optimization Regime. In Advances in Neural Information Processing Systems 37 (NeurIPS 2024), Datasets and Benchmarks Track. https://doi.org/10.52202/079017-2108
5. Hosseini, N., Sir, M., Jankowski, C., & Pasupathy, K. (2015). Surgical duration estimation via data mining and predictive modeling: A case study. AMIA Annual Symposium Proceedings, 2015, 640–648.
6. IBM. (2020). What are neural networks? Retrieved from https://www.ibm.com/cloud/learn/neural-networks Kadioglu, M. A., & Alatas, B. (2023). Enhancing Call Center Efficiency: Data Driven Workload Prediction and Workforce Optimization. The Eurasia Proceedings of Science Technology Engineering and Mathematics, 24, 96- 100. https://doi.org/10.55549/epstem.1406245
7. Keskin, N. Bora, & Zhang, C. (2024). Feature-based Scheduling and Dynamic Learning with a Large Backlog. SSRN Working Paper (SSRN 4852356). https://doi.org/10.2139/ssrn.4852356 Koçak, M., Calku, F., Gündaş, M., Poyraz, Z., Yazıcı, E., & Alakaş, H. M. (2022). Ekip çizelgeleme problemi: Filyasyon ekibi çizelgeleme. Journal of Turkish Operations Management, 6(2), 1314–1326. https://doi.org/10.56554/jtom.1101762
8. Loh, W. Y. (2011). Classification and regression trees. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 1(1), 14–23. https://doi.org/10.1002/widm.8

9. Mandi, J., Kotary, J., Berden, S., Mulamba, M., Bucarey, V., Guns, T., & Fioretto, F. (2024). Decision-Focused Learning: Foundations, State of the Art, Benchmark and Future Opportunities. Journal of Artificial Intelligence Research, 80, 1623–1701https://doi.org/10.1613/jair.1.15320
10. Öberg, J. (2020). Time prediction and process discovery of administration process. Digitala Vetenskapliga Arkivet. Retrieved from https://www.diva-portal.org/smash/get/diva2:1465971/FULLTEXT01.pdf (URL updated if needed—replace with correct link if different)
11. Ozpacaci, K., Bekli, S., & Kayisoglu, B. (2025). Incorporating Worker Heterogeneity in Flexible Flow Shop Environment. The Eurasia Proceedings of Science Technology Engineering and Mathematics, 34, 359-368. https://doi.org/10.55549/epstem.1755522
12. Pospieszny, P., Czarnacka-Chrobot, B., & Kobylinski, A. (2018). An effective approach for software project effort and duration estimation with machine learning algorithms. Journal of Systems and Software, 137, 184– 196. https://doi.org/10.1016/j.jss.2017.11.066
13. Ram, P. (2021). Generalized linear models: What does it mean? Great Learning. Retrieved from https://www.mygreatlearning.com/blog/generalized-linear-models/
14. Rosado-Galindo, A., & Dávila-Padilla, M. (2020). Tree-based missing value imputation using feature selection. Journal of Data Science, 18(4), 451–472. https://doi.org/10.6339/JDS.202010_18(4).0002
15. Ruiz-Chavez, Z., Salvador-Meneses, J., & Garcia-Rodriguez, J. (2018). Machine learning methods-based preprocessing to improve categorical data classification. In H. Yin, K. Meng, & Y. Gao (Eds.), Intelligent Data Engineering and Automated Learning – IDEAL 2018 (pp. 297–304). https://doi.org/10.1007/978-3-030-03493- 1_32
16. Sarı Ay, D. (2026). Evaluating Strategic Workforce Decisions in Aggregate Production Planning under Demand Uncertainty: A Two-Stage Stochastic MILP with Out-of-Sample Assessment. International Journal of Engineering Technologies IJET, 10(4), 69-80. https://doi.org/10.19072/ijet.1879062
17. SCIP Optimization Suite. (2022). SCIP: Solving constraint integer programs. Retrieved from https://www.scipopt.org
18. Soysal, S., Dengiz, B., & Atalay, K. (2021). Belirsizlik altında kaynak kısıtlı çok modlu çoklu proje çizelgeleme. Journal of Turkish Operations Management, 5(1), 598–614.
19. Şimşek, C., Dengiz, B., Karasakal, E., & İç, Y. T. (2022). Personel atama problemi: Bir kamu kurumu uygulaması. Journal of Turkish Operations Management, 6(2), 1297–1313. https://doi.org/10.56554/jtom.1128531
20. Torağay, O., & Pouya, S. (2023). A Monte Carlo simulation approach to the gap-time relationship in solving scheduling problem. Journal of Turkish Operations Management, 7(1), 1579–1590. https://doi.org/10.56554/jtom.1286288
21. Wilson, N. (1997). A genetic algorithm for the generalised assignment problem. Journal of the Operational Research Society. https://doi.org/10.1057/palgrave.jors.2600431
22. World Intellectual Property Organization (WIPO). (n.d.-a). Global Brand Database. Retrieved from https://www3.wipo.int/branddb/en/index.jsp
23. World Intellectual Property Organization (WIPO). (n.d.-b). WIPO International Classifications. Retrieved from https://www.wipo.int/classifications/en/
24. Zanakis, S. H. (1983). A staff-to-job assignment (partitioning) problem with multiple objectives. Computers & Operations Research, 10(4), 357–363. https://doi.org/10.1016/0305-0548(83)90014-2
25. Zulkepli, F. S., Ibrahim, R., & Saeed, F. (2017). Data preprocessing techniques for research performance analysis. In Advances in Intelligent Systems and Computing (pp. 157–162). https://doi.org/10.1007/978-981-10- 3779-5_20