Awake.AI is highlights predictive analytics as key to improving the quality of vessel scheduling which, it says, will change operational planning in maritime logistics for good.
In a recent article by Jussi Poikonen, VP of AI & Analytics, Awake.AI, the company focuses on how modern predictive analytics can help improve the quality of vessel scheduling data.
Poikonen focuses on vessel schedule prediction, specifically on predicting ‘normal’ ongoing voyages accurately using globally available sources such as Automatic Identification System (AIS) data, vessel particulars, weather data, etc.
He writes that many fairly common traffic scenarios fall outside of this scope, where vessels manoeuvre in ways which are not predictable using such data sources.
The article provides an outline of the approach Awake.AI takes to predicting routes and schedules of ongoing vessel voyages.
The company describes baseline and optimization models. The term baseline model is used for for rule-based models using non machine learning methods such as deterministic algorithms or heuristics to process data or perform predictions. Meanwhile, optimization models may use heuristics, probabilistic models, or black box machine learning (ML) models to make predictions.
In its modelling approach the company uses modular architecture. The modular consists of a sequence of submodels, which all affect the overall performance, but can be individually tuned and optimized as needed for any given target destination or customer use case.
To gain benefit from both baseline and ML-based optimization components, the submodels are constructed as model ensembles. In Awake.AI’s approach it means applying baseline and optimisation models for each prediction task in parallel, together with combiner models which aim to make optimal predictions based on the component inputs.
Awake.AI uses the article to demonstrate examples of specific scenarios, such as modeling location-dependent speed characteristics, where machine learning optimisation yields significant benefit in prediction accuracy.
Based on large scale performance comparisons with existing vessel schedule data sources, Poikonen said the company finds that the ML-based approach it is applying to ETA prediction provides a viable solution to common data quality problems in maritime logistics.