As recently as 2018, the United States Navy was actively considering adding seven years to the useful life of each ship in the fleet. With nearly 250 combat-ready vessels on the sea today – and hundreds more non-commissioned and support ships operating behind the scenes – that’s nearly a millennia of potential unplanned duty for the nation’s maritime forces.
Learning how to push the fleet further without putting sailors at excessive risk or sacrificing readiness is a data problem that experts are aiming to solve with GraphGrid.
Predictable maintenance is crucial for Navy vessels. It’s also a very complicated process involving systems and subsystems that use hundreds of thousands of components per ship, all with different requirements.
Some may only be serviced or upgraded when entirely out of the water – in what’s commonly known as “dry dock.” Others may be maintained in port while still others can be fine-tuned while operating at sea.
In this blog post, we’ll review the field experience of a system integrator working closely with Navy staff to develop safe maintenance schedules for an aging fleet, and how using GraphGrid to optimize the process could reduce risk, save money, and satisfy civilian government overseers in the process.
The challenge of eliminating waste and imprecision in maintenance scheduling
Maintaining a global fleet is as complicated as you might expect and involves multiple constituencies across the Navy Department, the broader Department of Defense, and various Congressional oversight committees.
Hard-won experience drives much of the initial decision-making, which makes sense. Experienced commanders know their ships. But manufacturers have a say too.
Makers of the systems and subsystems that keep vessels afloat prescribe their own recommended maintenance schedules, and they’re generally well-known to civilian overseers. Deviating from the recommended cadence – even if doing so makes sense to Navy leaders – may raise questions with those who control budgets.
Where does data fit? In the dark corners of what is essentially a negotiated settlement.
Experts and overseers generally default to spending as much as is needed to keep sailors safe while pushing the existing fleet to do more as shipyards slowly build and launch their replacements. It’s an inefficient process that favors contractors who want to sell more components and maintenance work to the Navy.
Looking for patterns in historical data
Aware of the potential for improvement, a systems integrator with extensive experience working with the Navy decided to look for hidden patterns buried in years of maintenance data. They selected GraphGrid to build a prototype, with the goal of presenting fleet command with more options for keeping vessels seaworthy.
Creating the prototype would take three months of data scientists, GraphGrid experts, Navy personnel, and the system integrator working together to build a knowledge graph in GraphGrid. Their goal: track the relationships and dependencies between systems, subsystems, and components.
With that level of understanding and rich history of maintenance data to analyze and test, the team could start asking questions.
For example: would it be possible to reduce the number of visits to dry dock maintenance? What would be reasonable yet still fall within an acceptable level of risk tolerance? Analysts wanted to see the potential for and impact of “knee in the curve” maintenance conducted when needed, but not before.
To get answers, they’d use the GraphGrid knowledge graph to examine years of service records from operating and decommissioned vessels. The team believed GraphGrid would surface real patterns of wear they could then compare with agreed-upon maintenance schedules. Every divergence would represent an opportunity to optimize.
Finding a 10% edge in the fight to keep the fleet afloat
Promising patterns revealed a number of potential opportunities. And yet the GraphGrid team chose to remain conservative, producing an analysis that showed just-in-time maintenance could reduce mission disruptions such as scheduled dry docks enough to account for a 10% reduction in annual maintenance spend.
It’s a lot for fleet commanders, Congress, and Defense Department leaders to consider as they continue to debate the options for optimizing maintenance.
In the meantime, the Navy has unveiled plans to increase its Sustained Maintenance Baseline – the annual spend on human resources, operations, and ship maintenance – from $55.5 billion to $130 billion in future years, as the fleet puts 50 new ships into service by 2049.
Reducing maintenance spending by 10%, as GraphGrid estimates, could increase agility, and readiness, and save billions for a global force that’s becoming more important by the day.
Get started now for free
On Sept. 15, 2021, GraphGrid unveiled the downloadable package in two editions: Enterprise and Ecommerce. Both editions include a full feature set. Enterprise edition includes Natural Language Processing (NLP) capabilities so you can turn text-based information into connected data. Ecommerce edition includes tools to create a smart shopping experience for customers and manage things like payment processing, invoicing, and order tracking.
The current 1.4 edition will be followed by a bi-annual release schedule. Current customers get 8 CPU cores, 32 GiB memory, and 1 GPU for all editions and their features and can be used through production for FREE.
Need more capacity or are you interested in having us run it in our cloud? Contact us now!
