Genetik Algoritmalar ile Deniz Taşımacılığında Hız Optimizasyonu

Öz

Gemicilik sektöründe yakıt tüketimini ve dolayısıyla sera gazı emisyonunu azaltmaya yönelik bazı düzenlemeler yapılmaktadır. Bunların başında seyir hızlarının düşürülmesi gelmektedir. Bu çalışmada, zaman kısıtlamalı genetik algoritmalar (GA) kullanılarak hız optimizasyonu çalışması yapılmıştır. Böylece gemi yolculuklarında en az yakıt tüketimi veren hız profilleri araştırılmıştır. Ham veri seti üzerinde çeşitli veri bilimi teknikleri uygulanarak eksik veya hatalı verilerin düzeltilmesi ve yerel anormallik faktör algoritması kullanılarak anormalliklerin temizlenmesi sağlanmıştır. Daha sonra, korelasyon analizi ile regresyon değişkenleri arasındaki ilişkiler belirlenmiştir. Bu bölümün sonunda da en az yakıt tüketimiyle sonuçlanan optimum gemi hızının araştırılması için ön işlemden geçirilmiş veriler kullanılarak yakıt tüketimi tahmin modeli ortaya çıkarılmıştır. Regresyon tahmin modeli için bir dizi makine öğrenmesi teknikleri kullanılmıştır. Bunlar; lineer regresyon(LR), K-En Yakın Komşuluk algoritması (KNN), Destek Vektörleri (SVR), Rastgele Orman (RF), ADABoost, Gradyan Artırmalar (GRB ve XGB) ve topluluk yöntemleridir. Ayrıca, tüm modeller için hiper parametre optimizasyonu yapılmıştır. Genetik algoritmada popülasyon bireyleri hız profilleri olup, ilklendirilmeleri için hız profillerinde rastgele modifikasyonlar denenmiştir. Sonuçta, genetik algoritmada farklı büyüklükteki mutasyon, çaprazlama, seçme, popülasyon sayısı, yolculuk gecikmesi ve hız limitleri kullanılarak minimum yakıt tüketimi ile sonuçlanan hız profilleri araştırılmıştır.

Anahtar Kelimeler

• Gemi hız optimizasyonu
• yakıt tüketimi
• genetik algoritma
• makine öğrenmesi
• regresyon modelleri

Voyage Speed Optimization Using Genetic Algorithm

Öz

Decreasing the fuel consumption and thus greenhouse gas emissions of vessels have emerged as a critical topic for both ship operators and policymakers in recent years. The speed of vessels has long been recognized to have the highest impact on fuel consumption. For this purposes, linear and non-linear methods such as KNN, support vectors, random forest, gradient boosting, ada boost, xgboost and ensemble models accuracy results are compared for better fuel consumption predictions in speed optimization using genetic algorithms. Furthermore, hyperparameters are researched for all models. The local outlier factor algorithm is used to eliminate outliers in prediction features. The overfitting problem is observed after hyperparameter tuning in boosting and tree-based regression prediction methods. The early stopping technique is applied for overfitted models. The aim of this study is to develop a speed optimization model using a time-constrained genetic algorithm (GA). The speed profiles are used as individuals in genetic algorithm and randomly initialization of speed profiles has been investigated. Also, various mutation, crossover, selection, population size, time of arrivals and speed limits have been tried to find a speed profile that results in minimum fuel consumption.

Anahtar Kelimeler

• Ship speed optimization
• fuel consumption
• genetic algorithm
• machine learning
• regression methods

Kaynakça

1. Aldous, L. G. (2016). Ship operational efficiency: performance models and uncertainty analysis UCL (University College London)].
2. Andrea, C., Luca, O., Francesco, B., & Davide, A. (2017). Vessels fuel consumption forecast and trim optimisation: A data analytics perspective. Ocean Engineering, 130, 351-370. https://doi.org/https://doi.org/10.1016/j.oceaneng.2016.11.058
3. Arslan, O., Besikci, E., & Olcer, A. (2014). Improving energy efficiency of ships through optimisation of ship operations. No. FY2014-3 IAMU.
4. Aydin, N., Lee, H., & Mansouri, S. A. (2017). Speed optimization and bunkering in liner shipping in the presence of uncertain service times and time windows at ports. European Journal of Operational Research, 259(1), 143-154. https://doi.org/10.1016/j.ejor.2016.10.002
5. Chaal, M. (2018). Ship operational performance modelling for voyage optimization through fuel consumption minimization World Maritime University]. Malmö, Sweden.
6. Changnan, W., Man, L., Songlin, Y., & Shasha, G. (2017, 2017/09). Optimization Analysis of USV Based on Genetic Algorithm. Proceedings of the 2017 5th International Conference on Mechatronics, Materials, Chemistry and Computer Engineering (ICMMCCE 2017).
7. Christos, G., Iraklis, L., & Gerasimos, T. (2019). Machine learning models for predicting ship main engine Fuel Oil Consumption: A comparative study. Ocean Engineering, 188, 106282. https://doi.org/https://doi.org/10.1016/j.oceaneng.2019.106282
8. De Andrade, L. C. O. (2014). Genetic Algorithms Application In Line Simplification. Faber, J., Nelissen, D., Hon, G., Wang, H., & Tsimplis, M. (2012). Regulated Slow Steaming in Maritime Transport. An assessment of options, costs and benefits.

9. förstudie initierad av Lighthouse, E. Consequences of speed reductions for ships.
10. Helong, W., Xiao, L., & Wengang, M. (2021). Voyage optimization combining genetic algorithm and dynamic programming for fuel/emissions reduction. Transportation Research Part D: Transport and Environment, 90, 102670. https://doi.org/10.1016/j.trd.2020.102670
11. Hu, Z., Zhou, T., Osman, M. T., Li, X., Jin, Y., & Zhen, R. (2021). A Novel Hybrid Fuel Consumption Prediction Model for Ocean-Going Container Ships Based on Sensor Data. Journal of Marine Science and Engineering, 9(4), 449. https://www.mdpi.com/2077-1312/9/4/449
12. IMO. (2016). Guidelines for the development of a Ship Energy Efficiency Management Plan (SEEMP). In RESOLUTION MEPC.282(70).
13. Karthe. (2016, 1/23/2022). Going Deeper into Regression Analysis with Assumptions, Plots & Solutions. Kee, K. K., Boung Yew, S. L., & Renco, K.-H. (2018). Prediction of Ship Fuel Consumption and Speed Curve by Using Statistical Method.
14. Kim, Y.-R., Jung, M., & Park, J.-B. (2021). Development of a Fuel Consumption Prediction Model Based on Machine Learning Using Ship In-Service Data. Journal of Marine Science and Engineering, 9(2), 137. https://www.mdpi.com/2077-1312/9/2/137
15. Psaraftis, H. (2019). Speed Optimization vs Speed Reduction: the Choice between Speed Limits and a Bunker Levy. Sustainability, 11, 2249. https://doi.org/10.3390/su11082249
16. Roar, A., Pierre, C., & Francois-Charles, W. (2020). Optimal ship speed and the cubic law revisited: Empirical evidence from an oil tanker fleet. Transportation Research Part E: Logistics and Transportation Review, 140, 101972. https://doi.org/10.1016/j.tre.2020.101972
17. Robert, N. (2022). Regression diagnostics: Testing the assumptions of linear regression. Retrieved 1/23/2022 from https://people.duke.edu/~rnau/testing.htm
18. Uyanık, T., Karatuğ, Ç., & Arslanoğlu, Y. (2020). Machine learning approach to ship fuel consumption: A case of container vessel. Transportation Research Part D: Transport and Environment, 84, 102389. https://doi.org/10.1016/j.trd.2020.102389
19. Yang, L., Chen, G., Zhao, J., & Rytter, N. G. M. (2020). Ship Speed Optimization Considering Ocean Currents to Enhance Environmental Sustainability in Maritime Shipping. Sustainability, 12(9), 3649. https://www.mdpi.com/2071-1050/12/9/3649