A research team from the Faculty of Economics at Universitas Indonesia has published an extensive analysis of container freight rate volatility covering the period January 2025 through June 2026. Using a dataset of over 180,000 spot and contract rates across 23 trade lanes, the study identifies structural shifts in volatility patterns that challenge conventional forecasting models used by the shipping industry.
The research identifies two novel volatility drivers that emerged during the study period. First, the rerouting of vessels around the Cape of Good Hope—a response to security concerns in the Red Sea—has introduced a new source of schedule uncertainty on Asia-Europe routes, with cascading effects on intra-Asia feeder services. Second, climate-related port disruptions, particularly flooding at major Southeast Asian transshipment hubs, have become a statistically significant predictor of rate spikes, a factor largely absent from pre-2024 forecasting models.
The hybrid model proposed by the researchers combines a long short-term memory neural network with a generalized autoregressive conditional heteroskedasticity framework, achieving a mean absolute percentage error of 8.7 percent on 30-day forecasts compared to 14.2 percent for the industry-standard Shanghai Containerized Freight Index futures. For Indonesian importers and exporters, the model's improved accuracy translates to more effective hedging strategies and better contract timing.
The study's practical implications extend beyond academic interest. The researchers have made their model available as an open-source Python package and are working with Indonesia's National Logistics Ecosystem team to integrate the forecasting tool into the government's trade information portal. If successful, this would give small and medium Indonesian exporters—who typically lack access to sophisticated freight rate analytics—a free tool for planning their shipping budgets and negotiating forwarder contracts.
Dr. Sari Nirmala
Associate Professor, Industrial Engineering
Dr. Sari Nirmala researches AI and machine learning applications in logistics at Institut Teknologi Bandung, with a focus on emerging economy contexts.
