Automatic Identification System data include the identification of the ship, its position, speed, draft and main dimensions. AIS data are commercially available from various providers, who also offer AIS-data derived information of different quality. Originally introduced for navigational safety, AIS data can be a powerful source of information for business intelligence applications.
Typical issues that may be addressed through AIS data mining and post-processing software are:
- How do partners/competitors run their networks? How many direct connections and transhipment do they offer?
- Which charter vessels have a higher chance of marine growth?
- Which ports/terminals have congestion issues?
- Do partners/competitors manage port operation faster? Why?
- Will the targeted berth be available on time?
- How do partners/competitors perform in terms of slow steaming and constant speed profile? How does this affect their fuel bill?
- How well are others performing regarding schedule integrity?
- What is the operational cost breakdown of other players?
- How much time do others spend in port and anchorage? How does this affect average speed?
- Which bunkering footprint do partners and competitors have?
- How efficiently do others bunker?
- Where do competitors dry-dock?
All these examples basically only require AIS data, an electronic map and a set of GPS (Global Positioning System) coordinates to identify key areas (ports, quays, emission control areas, etc.). But an even more interesting insight can be gained from combining AIS data with other data, e.g. data on fuel consumption, emissions, weather and sea state, granular geospatial objects or ship schedules. And such insight can support maritime business at operational, tactical or strategic level:
Business intelligence helps companies to take faster and smarter decisions. “Companies” in this sense may be different stakeholders in the maritime industry such as ship operators and owners, port operators and authorities, insurance companies, commodity traders or maritime service providers. In order to generate the business intelligence, AIS data is merged with other data sources to extract new insight. With billions of data records per year and terabytes of data (“Big Data”), this demands powerful data warehousing and processing capabilities.
A few concrete examples illustrate where our industries are heading:
1. DNV GL was tasked with emission monitoring for of the Norwegian NOX(subscpt x] fund. Initially, only the coastal traffic in Norwegian territorial waters was monitored via AIS to determine the emissions for each ship. Now it is possible to establish emission inventories for any part of the world. Similarly, AIS data may be used to create accurate proxies on (daily) fuel consumption for a whole range of vessels. This is used, e.g., in DNV GL’s “ECO Insight” tool to verify noon report data and to compare average fuel consumption against similar vessels of competitors on similar routes.
2. Delay management in container shipping: In container shipping, delays are the rule rather than the exception. A delay can become costly as berths become unavailable but speeding up burns more fuel and cargo from skipped ports then needs to be repositioned. Hence a good delay management system is important. An operations’ department manager would benefit from a review of where his services did not run on schedule (root cause analysis) and what mitigation actions were taken (speed up, skip port, cut and run, etc.). With this information, he can discuss mitigating costs with the sales department, which is interested in keeping the ship on course to its original port call schedule to avoid customer complaints. By combining schedule data with vessel port call/anchorage time data, we can pinpoint where operations went wrong and with speed data we can see where vessels sped up to catch up again. AIS data even allow a peek into competing services to see how competitors manage to keep their ships on schedule.
3. Voyage management AIS data can be used to analyse your own and competitors’ performance regarding voyage management. As an example, AIS gives speed information with a much higher data density than noon reports. Detailed AIS data can be used to derive various key indicators:
- Speed variability during deep sea passage: A too high starting speed, while slowing down later on to match the arrival time in port indicates poor voyage planning. Speed variations around a constant mean may be due to weather.
- Average sea passage speed: A too high pro-forma speed may indicate poor pro-forma scheduling.
- Operating profile standard deviation (head-haul/back-haul).
- Port times: Long port stays may indicate poor port productivity or poor coordination with terminal operator.
- Anchorage times: Anchorage times due to early arrival may indicate poor voyage planning.
- Nautical miles/day.
For one client, irregular speed patterns resulted in ~2 mUSD higher fuel bill per year for an 8,500 TEU containership. Table II gives three ship operators (made anonymous as A, B, and C), illustrating the quantitative insight harvested from our AIS analyses. In a subsequent root-cause analysis, the specific ship was tracked looking at mooring times, port times and transit times.
Table II: AIS data analysis for overall operations for three ship operators A, B, and C on same trade
|Time share sea passage||69.5%||71.5%||70.0%|
|Time share maneuvering||5.0%||5.4%||5.4%|
|Time share anchoring||7.4%||6.0%||6.0%|
|Time share mooring||18.1%||17.1%||17.8%|
|Average sea passage speed||16.5kn||16.0kn||16.1kn|
|Standard deviation in speed||2.78kn||2.45kn||2.81kn|
|Nautical miles / day||291nm||291nm||285nm|
There are many more potential applications for AIS-based business intelligence, but these examples may suffice to realise the scope and power of this new technology.