“Data” has become quite the buzzword in the maritime industry, and for good reason. We make sense of the world, and make smarter decisions, by collecting and interpreting data. With the increasing complexity, competitiveness, and regulation of maritime commerce, intelligent use of data is the best bet to stay efficient, profitable, and compliant. 

Whatever your involvement in the trading and movement of seaborne cargos, data and the insights you derive from them play a critical role in your success. In this article, I explore the importance of data in our industry, what it means to have “good data,” and the role of data science in shipping, both in general and at Veson.

The scale of maritime operations

In recent years, our industry has seen a paradigm shift from using limited individual data sets to using broader connected data ecosystems. In 2023 alone, our clients on the Veson IMOS Platform made 808,000 port calls, moved 6 billion tons of freight under contracts worth $170 billion, spent $29 billion on bunker fuels, and incurred $11 billion in demurrage costs. With figures this large, even tiny improvements in operational efficiency can generate large financial savings. But how can we drive improvements in such complex and interdependent operations? One answer is data.   

Many of our clients have come to realize that the difference between industry leaders and laggards will increasingly be defined by how well they harness data. This extends beyond commercial gains to encompass regulatory and environmental objectives. The International Maritime Organization (IMO) aims for global shipping to achieve 40% reduction in carbon intensity, 5-10% adoption of zero-emission fuels, and 20-30% reduction in greenhouse gas emissions by the end of this decade. It has set our industry the goal of net-zero emissions by 2050. Data will be instrumental in meeting these ambitious targets, and in exploiting the opportunities that always come with new challenges.

Undoubtedly, we have a “big data” challenge facing maritime. The good news is that new technologies can help. We’ll discuss those technologies later, but first let’s align on what makes data “good data.”

Types and characteristics of useful data

Data, in essence, are known facts about the world. They are distinct from estimates and predictions, which are often derived from data and have an element of uncertainty. Within our industry, we deal with two main types of data: internal data and market data. Internal data comprise information about your company’s commercial and technical operations, while market data are facts generated by the entire maritime industry. Examples of market data include the particulars and movements of individual vessels, at the most zoomed-in level, to regional congestion and global commodity flows, at the most zoomed-out.

Good data are characterized by several important qualities:

  • Relevant: Data must relate to the decision at hand.
  • Timely: Data must be available before the point of decision.
  • Trustworthy: Data must be reliable, whether generated internally or sourced from suppliers.
  • Validated: Data validated across multiple sources will be typically more reliable than data from a single source.
  • Standardized: Agreed data formats facilitate communication and collaboration.
  • Consumable: Delivery mechanisms should make data easy to consume and action.

Large datasets, even of good data, can be difficult to manage. At Veson, we see several innovative ways of confronting this challenge: from structuring internal and external data to align with workflows, to democratizing data access through natural language querying, to parsing unstructured text data into structured and contextual datasets.

The role of data science

Data science is integral to extracting meaningful insights from large datasets. At its core, data science involves building models to derive unknown information from known data. Models must balance complexity and simplicity, separating robust signals from unpredictable noise.

Leveraging Data And Data Science In Shipping For Better Decisions Data Model Screens

By melding modern computing power with statistical expertise, we can create insights from data that align with workflows, enable rapid data access through natural language, and convert unstructured data into structured and scalable datasets.

Here at Veson, our data science team works hard to turn data into insights that help our clients make faster and smarter decisions. It does this in various ways: by validating and cleaning raw collected data; by using statistical methods to detect patterns and clusters in cleaned data; by building mathematical models to generate estimates and predictions consistent with data; and by deploying modern machine learning and artificial intelligence algorithms, with care and rigor, to maritime commercial problems.

Examples of data-driven decisions in maritime

Data science in shipping can inform a broad range of topics. From macroeconomic trading strategies based on global commodity flows to voyage estimation and cost control based on port congestion, data science applications in maritime decision-making are manifold. For instance, technical specifications of vessels can inform risk assessments related to sanctions, piracy, and war zones. By integrating external data with internal operations, businesses can make more informed decisions about routing, re-routing, and overall fleet management.

In financial contexts, our vessel valuation models are the industry standard for commercial loan monitoring and analysis. When adjusted to remove the effects of depreciation, they provide pure market signals about the changing value of tonnage. Model outputs like these help investors, lenders, and analysts understand market cycles and evaluate large strategic investment decisions.

Leveraging Data And Data Science In Shipping For Better Decisions Vesselsvalue Screen

VesselsValue vessel valuation example, fixed age value

To support sustainability efforts, we can also predict fuel consumption, carbon emissions, and other operational metrics. Veson data scientists have built a physics-based statistical model which estimates fuel consumption from real-time data of the movements of vessels, along with technical data about their fuel efficiency. This model feeds into various products that help our clients manage their environmental regulatory compliance and their emissions trading exposures.

Leveraging Data And Data Science In Shipping For Better Decisions Cii Screen

How data science comes to life at Veson

At Veson, we offer a suite of products designed to harness the power of data. Our Oceanbolt platform, for instance, takes AIS data about vessel positions and times and transforms them into actionable insights about port calls, voyages, cargo movements, and global commodity flows. By integrating AIS data, vessel dimensions, and engine performance characteristics, we can create models that estimate annual voyage and period-based emissions for any vessel. This allows for fleet comparisons and optimization, helping clients reduce their environmental impact and operational costs. Such high-quality data equip you with what you need to make informed decisions.

Leveraging Data And Data Science In Shipping For Better Decisions Oceanbolt Screen

Oceanbolt port congestion data

Other products, such as our position lists, spot charters, and commodity transaction databases, further enhance our data offerings. These tools provide critical information for managing pre-fixture processes, compliance, energy efficiency, and more, giving you a comprehensive view of vessel and fleet operations. Leveraging data and data science is crucial for optimizing maritime operations. By integrating high-quality data into decision-making processes, and by using advanced data science techniques, we can help our clients achieve both commercial success and regulatory compliance. As we continue to innovate and refine our data offerings, we remain committed to empowering the maritime industry with the tools and insights it needs to navigate the future confidently.