Predictive modeling is playing an increasingly important role in improving safety across the maritime industry. Even if you’re not sure what it is, you’ve probably heard of at least one company that uses predictive models to predict potential risks to their fleets. So what is predictive modeling and how much value is in it?
Let’s start with what predictive modeling isn’t. While it enables companies to give early warning of their key risks, predictive modeling is not a modern method of divination. It does not physically enable anyone to see into the future, and it does not include a crystal ball.
Predictive modeling is a lot easier than many think – it’s a combination of math and statistics that is used to calculate potential risks from in-house data. When done correctly, predictive modeling has been shown to prevent preventable incidents, save lives at sea and prevent damage to ships, the environment and corporate profits. If predictive modeling is ineffective, it’s a lost opportunity for everyone in the industry.
To make predictive modeling effective, it takes a leap of faith to trust the analysis, even if it is contrary to what organizations might expect. Because predictive modeling treats data differently, it reveals problems that have not been observed with other systems. It gives the marine industry a better chance of anticipating problems, and every disaster avoided means safer seas for all.
Predictive modeling in other industries
Companies around the world share their data for predictive models and benefit from tailored risk reports that reveal the most likely threats to their crew. Predictive modeling has also proven itself in many other industries.
After the 1999 Ladbroke Grove rail accident, the rail industry established the Rail Safety Standards Board (RSSB) to implement effective data exchange. The incident was caused by signal visibility issues that had been reported multiple times but not resolved. The rail industry saw the need to create a central safety database to prevent future incidents.
The RSSB collects data from every UK railway company and uses it to achieve safer practice across the industry. The RSSB identifies the causes of serious disruptions and equips railway companies with the resources they need to reduce risk. As a result, the accident rates have dropped significantly.
In the US, the use of predictive modeling has been shown to be extremely effective in reducing crime and improving safety in cities. For example, the Richmond, Virginia police used historical data to predict when guns would be fired on New Year’s Eve 2003. They used the predictions to plan their surveillance routes and saw a 47 percent decrease in gunfire. The use of predictive modeling saved the troop $ 15,000 and was viewed as a resounding success.
How changing attitudes lead to safer seas
The maritime industry benefits from an effective data exchange in a way that has never been seen before. Historically, data exchange in the maritime industry has always been reactive. The broader industry would be privy to basic details of an incident through headlines or investigations when it was too late. This meant the data exchanged was extremely simple and the broader industry had limited opportunities to learn from the information. This method also meant that, unfortunately, by the time information became available, it was too late to avoid an incident at sea.
In the past, shipping companies have been reluctant to share their data because they feared reprisals for mistakes and would lose ground in a competitive market. As attitudes evolve and companies embrace the idea of ââsharing their data for the common good, predictive modeling will gradually become more robust and accurate for anyone who depends on protecting their crews at sea.
Why is predictive modeling going wrong?
Unsuccessful predictive modeling is always due to one factor – insufficient data. Predictive modeling is only as good as the data on which it is based. It is an unfortunate fact that organizations that run predictive models on only publicly available baseline data do not see effective reports with valuable insights.
Over the years, historical analyzes have been created using data that companies are required to publish, so that minor details that can lead to incidents at sea may be overlooked. This makes it easy for companies to get their data wrong and put them at risk. If the data is incorrect, predictive modeling cannot highlight the events that are most likely to endanger a ship or its crew.
To function well, predictive modeling requires internal data beyond what is available in the company’s incident management system. It is critical to collect internal data from each and every data set so that the process can analyze every possible risk within a fleet. Effective predictive modeling means gathering information from every data set, source, and owner within an organization.
How to do predictive modeling right
In order for predictive modeling to work, we need to look at the small events that can turn out to be leading indicators. For predictive modeling to be really effective and effective, we need to analyze data from different locations and from each dataset.
The sharing of data from each dataset enables an accurate analysis of every possible situation. Analyzing data from each dataset makes it much better possible to pick up on the leading incidents that would otherwise be overlooked until an incident occurs.
The more companies share their data for predictive modeling, the more we create a well-rounded picture of what exactly is going on in the maritime industry and provide a bird’s eye view to specialized risk management companies doing predictive modeling. This unique perspective enables greater accuracy in calculating risks and safer seas for everyone.
More and more maritime companies are doing predictive modeling with the aim of improving safety for everyone. Attitudes towards data exchange have also changed in recent years and companies have become more open to the idea. The good news is, the more companies share their data, the closer we will get to eliminating all avoidable incidents in the maritime sector.
HiLo Maritime Risk Management uses predictive big data analytics to improve seafarers’ safety and identify areas of risk that are overlooked by other systems.
The opinions expressed here are those of the author and not necessarily those of The Maritime Executive.