Dijital Dönüşüm Sürecinde Akıllı Üretim Planlama: Yapay Sinir Ağları ve Genetik Algoritma Bütünleşik Yaklaşımı

Öz

Günümüzün yoğun rekabet ortamında üretim planlama süreçlerinin etkinliği, işletmelerin sürdürülebilirliği ve operasyonel verimliliği açısından kritik bir öneme sahiptir. Bu çalışmada, İstanbul’da faaliyet gösteren orta ölçekli bir makine üretim tesisinde üretim planlama kararlarını iyileştirmek amacıyla Yapay Sinir Ağları (YSA) ve Genetik Algoritma (GA) yöntemlerini bütünleştiren hibrit bir yapay zeka modeli geliştirilmiştir. Önerilen model iki aşamadan oluşmaktadır: ilk aşamada, sipariş adedi, ürün tipi, stok miktarı ve iş gücü kullanımı değişkenlerine dayalı olarak üretim süresi YSA modeli ile tahmin edilmiştir (R² = 0.91, MAPE = %8.7). İkinci aşamada ise, elde edilen tahmin değerleri GA tabanlı optimizasyonda kullanılarak siparişlerin üç üretim hattına kapasite kısıtları altında dengeli biçimde atanması sağlanmıştır. Evrimsel operatörler (seçim, çaprazlama, mutasyon) aracılığıyla çalışan GA modeli, iş yükü dengesizliğini minimize ederek genel planlama verimliliğini artırmıştır. Karşılaştırmalı analizler, önerilen yaklaşımın manuel planlamaya göre ortalama üretim süresini %18.6, teslimat gecikmelerini %45.6 ve toplam işçilik maliyetini %11.9 oranında iyileştirdiğini göstermektedir. Sonuç olarak, geliştirilen YSA–GA tabanlı model, küçük ve orta ölçekli işletmeler için düşük maliyetli, uygulanabilir ve dijital dönüşüm sürecine bir karar destek aracı olarak değerlendirilmektedir.

Anahtar Kelimeler

• Üretim Planlama
• Yapay Sinir Ağları
• Genetik Algoritma
• Optimizasyon
• Akıllı Üretim

Intelligent Production Planning in the Digital Transformation Process: An Integrated Approach of Artificial Neural Networks and Genetic Algorithm

Öz

In today’s highly competitive industrial environment, the efficiency of production planning processes plays a critical role in ensuring the sustainability and operational performance of enterprises. This study develops a hybrid artificial intelligence model that integrates Artificial Neural Networks (ANN) and Genetic Algorithms (GA) to enhance production planning decisions in a medium-sized machine manufacturing company operating in Istanbul. The proposed model consists of two stages: in the first stage, production time is predicted using an ANN model based on variables such as order quantity, product type, stock level, and labor utilization (R² = 0.91, MAPE = 8.7%). In the second stage, the predicted values are utilized in a GA-based optimization process to achieve a balanced assignment of orders to three production lines under capacity constraints. The GA model, operating through evolutionary operators (selection, crossover, and mutation), minimizes workload imbalance and improves overall scheduling efficiency. Comparative analyses indicate that the proposed approach reduces average production time by 18.6%, delivery delays by 45.6%, and total labor costs by 11.9% compared to the manual planning method. Consequently, the developed ANN–GA-based model is evaluated as a low-cost, applicable decision-support tool that contributes to the digital transformation process of small and medium-sized enterprises (SMEs).

Anahtar Kelimeler

• Production planning
• Artificial Neural Networks
• Genetic Algorithm
• Optimization
• Intelligent Manufacturing

Kaynakça

1. Akyol DE., Bayhan GM. A review on evolution of production scheduling with neural networks. Computers & Industrial Engineering 2007; 53(1): 95-122.
2. Bishop CM. Pattern recognition and machine learning. New York, NY: Springer 2006.
3. Bubak A., Rolf B., Reggelin T., Lang S., Stuckenschmidt H. An LSTM network-based genetic algorithm for integrated procurement and scheduling optimisation. International Journal of Production Research 2025; 63(11): 4036-4065.
4. Del Gallo M., Mazzuto G., Ciarapica FE, Bevilacqua M. Artificial intelligence to solve production scheduling problems in real industrial settings: A systematic literature review. Electronics 2023; 12(23): 4732.
5. Gen M., Cheng R. Genetic algorithms and engineering optimization. Hoboken, NJ: John Wiley & Sons 2020. Georgy M., Basily SY. Using genetic algorithms in optimizing construction material delivery schedules. Construction Innovation 2008; 8(1): 23-45.
6. Goldberg DE. Genetic algorithms in search, optimization, and machine learning. Reading, MA: Addison-Wesley 1989.
7. Haykin S. Neural networks: A comprehensive foundation (2nd ed.). Upper Saddle River, NJ: Prentice Hall 1999. Huang CL.The construction of production performance prediction system for semiconductor manufacturing with artificial neural networks. International Journal of Production Research 1999; 37(6): 1387-1402.
8. Kamble SS., Gunasekaran A., Sharma R. Analysis of the driving and dependence power of Industry 4.0 technologies in smart manufacturing. Technological Forecasting and Social Change 2023; 187: 122211.

9. Li J., Wang Y., Zhang X. A hybrid artificial neural network and genetic algorithm approach for production scheduling under capacity constraints. Expert Systems with Applications 2023; 213: 118935.
10. Mahmud S. Modelling and optimizing supply chain integrated production scheduling problems. University of New South Wales (Australia) 2023.
11. Ojstersek R., Brezocnik M., Buchmeister B. Multi-objective optimization of production scheduling with evolutionary computation: A review. International Journal of Industrial Engineering Computations 2020; 11(3): 359-376.
12. Pinedo M. Scheduling: Theory, algorithms, and systems. New York, NY: Springer 2016.
13. Roeva O., Zoteva D., Roeva G., Ignatova M., Lyubenova V. An effective hybrid metaheuristic approach based on the genetic algorithm. Mathematics 2024; 12(23): 3815.
14. Wang L., Liu H., Xia M., Wang Y., Li M. A machine learning-based EMA-DCPM algorithm for production scheduling. Scientific Reports 2024; 14(1): 20810.
15. Zhang F., Nguyen S., Mei Y., Zhang M. Genetic programming for production scheduling. Singapore: Springer 2021.