KARABİGA LİMAN’NDA ELLEÇLENEN YÜKLERİN ZAMAN SERİSİ ANALİZİ VE TAHMİNİ

Yıl 2026, Cilt: 5 Sayı: 1, 99 - 117, 31.01.2026

Elvan Deniz

https://izlik.org/JA88RA74PW

Öz

Marmara Denizi’nin güney kıyısında yer alan Karabiga Limanı, sanayi kümelenmelerine yakınlığı ve genel kargo, kuru dökme ve sıvı dökme yüklerdeki çeşitlendirilmiş elleçleme kabiliyeti sayesinde Türkiye’nin bölgesel lojistik sistemi içinde stratejik bir önem kazanmıştır. Bu çalışma, resmi liman istatistiklerini kullanarak Karabiga Limanı’nın 2012–2024 dönemine ait yıllık toplam yük elleçleme hacimlerini analiz etmekte ve 2025–2027 dönemi için kısa vadeli tahminler üretmektedir. Minitab 17 yazılımı kullanılarak doğrusal, kuadratik (parabolik) ve üssel trend modelleri; hareketli ortalamalar (3 ve 5 terimli) ile basit ve çift üssel düzeltme yöntemleri uygulanmıştır. Modellerin performansı Ortalama Mutlak Yüzde Hata (MAPE), Ortalama Mutlak Sapma (MAD) ve Ortalama Kare Sapma (MSD) ölçütleriyle değerlendirilmiştir. Bulgular, uzun dönemde genel bir büyüme eğilimi bulunduğunu; 2018 sonrası belirgin dalgalanmalar ve son yıllarda ise hafif bir yavaşlama/gerileme işaretleri görüldüğünü ortaya koymaktadır. Rakip modeller arasında kuadratik trend modeli, açıklayıcılık ile tahmin doğruluğu arasında en güçlü dengeyi sağlayarak yük artışının doygunluk noktasına yaklaşabileceğine işaret etmektedir. Sonuç olarak, önerilen tahminleme çerçevesi; kapasite planlaması, altyapı yatırımının değerlendirilmesi ve orta ölçekli bölgesel limanlarda operasyonel yönetim için pratik bir karar destek aracı sunmaktadır.

Anahtar Kelimeler

Karabiga limanı , zaman serisi analizi , talep tahmini

TIME SERIES ANALYSIS AND FORECASTING OF CARGO HANDLING AT THE PORT OF KARABİGA

https://izlik.org/JA88RA74PW

Öz

Ports are critical nodes in global supply chains, and reliable forecasting of cargo throughput is essential for aligning terminal capacity, equipment investment, workforce planning, and service reliability with anticipated demand. In developing logistics regions, medium-scale ports often experience rapid structural changes driven by local industry dynamics, infrastructure expansion, and evolving trade patterns. The Port of Karabiga, situated on the southern coast of the Sea of Marmara, exemplifies this context: it serves nearby industrial facilities and Organized Industrial Zones (OIZs), provides multi-cargo services (general cargo, dry bulk, liquid bulk), and increasingly functions as a regional alternative to neighboring ports. Within this setting, quantifying historical handling patterns and generating evidence-based short-term forecasts can enhance both public and private decision-making.
This study examines the Port of Karabiga’s annual total cargo handling volumes over the period 2012–2024 and develops forecasts for 2025–2027 using classical time-series techniques. The dataset consists of complete annual observations obtained from official sector statistics and is suitable for non-seasonal modeling due to its yearly frequency. Prior to model estimation, distributional adequacy was checked via the Anderson–Darling normality test, indicating that the data do not violate normality assumptions at conventional significance levels. Given the limited number of annual observations and the absence of an explicit seasonal component, Box–Jenkins (ARIMA) modeling was not prioritized; instead, the analysis focused on trend-based and smoothing-based methods that are widely applied for annual port throughput series.
Modeling was conducted in Minitab 17 using linear trend analysis to capture steady growth, quadratic trend analysis to represent accelerating or decelerating growth and potential saturation, and exponential trend analysis to reflect compound growth patterns. In addition, 3- and 5-period moving averages were applied to smooth short-term volatility, while single and double exponential smoothing (Holt-type) models were used to represent adaptively updated level and trend components. To ensure an objective comparison across competing specifications, forecasting accuracy was evaluated using multiple error metrics: Mean Absolute Percentage Error (MAPE) for scale-free interpretability, Mean Absolute Deviation (MAD) for average absolute error in the original units, and Mean Squared Deviation (MSD) to reflect sensitivity to larger deviations.
Empirically, the port’s long-term trajectory reflects growth over the full period, with a marked rise through the mid-to-late 2010s followed by more pronounced fluctuations after 2018 and a slight weakening in some recent years. Among all candidate models, the quadratic trend specification achieved the best overall performance, combining relatively high explanatory power with the lowest error indicators, and implying a decelerating growth profile consistent with emerging capacity and market constraints. Diagnostic evidence also suggested mild positive autocorrelation in the quadratic trend residuals; applying an AR(1) correction and subsequent Ljung–Box testing supported the adequacy of residual independence for inference. The resulting forecasts for 2025–2027 provide a quantitative reference for operational and strategic planning, while highlighting that alternative models (e.g., exponential trend or smoothing methods) may yield differing scenarios and should be used for sensitivity analysis.
Overall, this research contributes to the limited empirical evidence on medium-scale Turkish ports by demonstrating how a multi-model, error-metric-driven approach can generate actionable forecasts and support capacity planning, investment timing, and performance management under uncertainty.

Anahtar Kelimeler

Port of Karabiga , time series analysis , demand forecasting

Kaynakça

• Akkan, B., & Çalışır, V. (2022). Klasik zaman serisi yöntemleri ile konteyner elleçleme tahmini. Journal of Anatolian Environmental and Animal Sciences, 7(3), 341-349. https://doi.org/10.35229/jaes.1133335
• Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: forecasting and control. John Wiley & Sons.
• Brown, R. G. (1959). Statistical forecasting for inventory control. McGraw-Hill.
• Brown, R. G. (1963). Smoothing, forecasting and prediction of discrete time series. Prentice-Hall.
• Chatfield, C. (2000). Time-series forecasting. Chapman and Hall/CRC.
• Chatfield, C., & Xing, H. (2019). The analysis of time series: An introduction with R. Chapman and hall/CRC.
• Deniz, E. (2025). Marmara Bölgesinde Elleçlenen Yüklerin Analizi ve Tahmin Edilmesi. Trakya Üniversitesi Mühendislik Bilimleri Dergisi, 26(1), 1-15. https://dergipark.org.tr/tr/pub/tujes/article/1700210
• Gardner Jr, E. S. (1985). Exponential smoothing: The state of the art. Journal of forecasting, 4(1), 1-28. https://doi.org/10.1002/for.3980040103
• Goss, R. O. (1990). Economic policies and seaports: Are port authorities necessary?. Maritime Policy & Management, 17(4), 257-271. https://doi.org/10.1080/03088839000000032
• Hyndman, R.J., & Athanasopoulos, G. (2021) Forecasting: principles and practice, 3rd edition, OTexts: https://otexts.com/fpp3
• Hyndman, R. J., & Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4), 679–688. https://doi.org/10.1016/j.ijforecast.2006.03.001 (Erişim tarihi: 17 Ekim 2025).
• Jugović, A., Hess, S., & Poletan Jugović, T. (2011). Traffic demand forecasting for port services. Promet-Traffic&Transportation, 23(1), 59-69. https://doi.org/10.7307/ptt.v23i1.149
• Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (1998). Forecasting: Methods and Applications (3rd ed.). John Wiley & Sons.
• Montgomery, D. C., Jennings, C. L., & Kulahci, M. (2015). Introduction to time series analysis and forecasting. John Wiley & Sons.
• Notteboom, T. E., & Rodrigue, J. P. (2005). Port regionalization: Towards a new phase in port development. Maritime Policy & Management, 32(3), 297-313. https://doi.org/10.1080/03088830500139885
• Öztemiz, H., & Vatansever, K. (2023). Türkiye küresel konteyner liman projeleri: 2035 yılı konteyner trafik hacmi ve dış ticaret projeksiyonu. Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 24(3), 261-298. https://doi.org/10.53443/anadoluibfd.1253057
• Shankar, S., Ilavarasan, P. V., Punia, S., & Singh, S. P. (2020). Forecasting container throughput with long short-term memory networks. Industrial management & data systems, 120(3), 425-441. https://doi.org/10.1108/IMDS-07-2019-0370
• Tan, N. D., Yu, H. C., Long, L. N. B., & You, S. S. (2021). Time series forecasting for port throughput using recurrent neural network algorithm. Journal of International Maritime Safety, Environmental Affairs, and Shipping, 5(4), 175-183. https://doi.org/10.1080/25725084.2021.2014245
• Tuncel, G., & Deniz, E. (2024). Optimization of container terminal operations using response surface methodology. International Journal of Simulation Modelling (IJSIMM), 23(4), 563-574. https://doi.org/10.2507/IJSIMM23-4-695
• Türkiye Liman İşletmecileri Derneği. (TÜRKLİM) (2023). Liman istatistikleri ve sektörel raporlar. (Erişim tarihi: 17 Ekim 2025). https://www.turklim.org
• Türkiye Liman İşletmecileri Derneği (TÜRKLİM). (2025). Sektör istatistikleri. (Erişim tarihi: 17 Ekim 2025). https://www.turklim.org/sektor-istatistikleri/
• UN Trade and Development (UNCTAD). (2024). Review of maritime transport 2024: Navigating maritime chokepoints. United Nations. https://unctad.org/system/files/official-document/rmt2024_en.pdf (Erişim tarihi: 17 Ekim 2025).