WASHINGTON — The year 2017 was among the worst in recent memory for the U.S. Navy.

A string of accidents in the Pacific, including two deadly collisions with commercial ships, left 17 sailors dead and a mountain of questions about how the Navy was using its Japan-based forces, and whether the service as a whole was overwhelmed by demands and responsibilities.

The accidents led to wholesale changes in leadership for the 7th Fleet based in the region, including a new commander, Vice Adm. Phil Sawyer, who was tasked with returning his fleet to a high state of readiness in the world’s most dangerous seas.

Sawyer has been in command since August 2017, and since then he’s led a shift in how 7th Fleet deploys its ships, a radical departure from the system in place prior to his arrival. In the past, ships relied on revolving certifications in their essential tasks, certifications that were lapsing routinely by 2017.

Defense News caught up with Sawyer, who was afloat on his command ship Blue Ridge in the East China Sea, to discuss the changes and what’s different about the 7th Fleet today.

The Navy made the decision to repair the destroyer John S. McCain in Japan rather than return it to the U.S. for refitting. It’s back in the water now. When will it go back on patrol?

I expect them to complete the [repairs] this spring. And then once they complete that, then they’ll go through sea trials and all the things we do when ships come out of deep maintenance like she’s gone through.

When we spoke last, you were implementing a plan known as “Optimized Fleet Response Plan—Japan” to institute a training cycle closer to the one U.S.-based ships go through to prepare for deployments. How’s that going?

When I talked to you last it was a two-year cycle and it was based on continuous certifications. That has since changed, and it is now a 36-month cycle, and that balances maintenance, training and operations, and it very closely resembles the models they use in the continental United States. We are in run with that new model, we haven’t had anyone complete the full 36-month cycle yet. But once they do complete it, it feeds into a maintenance period, and then they start the model over again.

You’ve gone from a model where ships, considered permanently deployed, must maintain their certifications based on periodicity. Now they earn certifications as they work up to deployments the way they do stateside. How is that adjustment going?

I’ll highlight two things: We have basic phase, advanced and integrated phases. All of those phases use the exact same standards used in the rest of the Navy. So I think when you talk to the guys out here on the waterfront, they recognize and appreciate the time given for training and certification. They know that it’s cordoned off and will know with predictability when they are going to be training, whether its basic or advanced. This gives the teams, the trainers, the families — everybody — the ability to plan.

So I think in general it’s producing a better force, and that’s key. But I think the sailors appreciate the predictability.

Does moving to this kind of model reduce the amount of days underway that are available to you from Japan-based ships?

That’s a complicated question. If you go down to the waterfront in Norfolk or San Diego, you’ll have 10 or 15 ships out there on the pier, and they are all going to be in different phases of their model. The model in Yokosuka broadly includes maintenance, training and certification, and then operations. So when you go down to the waterfront here you are going to find one-third of my ships in maintenance, one-third in training and one-third in operations. That is the sustainable model that goes to the future.

Each of the units entered the cycle in a different place, but this is about long-term sustainability.

There was concern after the accidents that the demands on your ships were too great. Does this put you in a better place?

If you go back to the reports in the wake of the accidents, one of the things that stands out was that ships were not getting the training and certification needed, and the maintenance they needed was getting shorted. And the reason was operations. So we don’t do that anymore. We recognize we have to do a certain level of training and certification; you have to do a certain level of maintenance to be long-term sustainable. That is what we are doing now.

Regardless of what they were supplying back then, now we’re at the point where we make sure the ships are maintained and they are trained and certified before they go out and do operations.

To the larger point, there is always more demand than I have supply. And that has been the case since I first started in the Navy, and it’s always going to be the case. I only supply that which I can supply. And said another way: We’re not going to put them out there to do operations unless they are trained and certified. It’s critical to our force-generation model, and I’m allegiant to making sure that remains the standard.

How do you assess the progress you’ve made in the past year implementing changes in 7th Fleet?

We’ve made a lot of progress. We still have some work to do. As I tell my team when they ask how I assess it, I say: “I think we’re fair in the channel and we are gaining speed.”

And as I told you back then, the sailors we lost and their families that provide us the constant motivation, we owe it to them, and we owe it to ourselves to get it right. Because if we don’t get it right, we wont be able to operate safely or confidently. Proper readiness, proper preparation enables safe operations. They are inseparable.

In the wake of the collisions, there was a directive put out by the commander of Naval Surface Forces Pacific that all the surface ships need to get on a circadian rhythm watch bill to get sailors more sleep. I’ve heard anecdotally that there has been some uneven compliance with that directive. How are you seeing it play out in 7th Fleet?

I think there are really two questions in there, and one of them is manning. We think of manning in terms of fit (is the right person there) and fill (do they have enough). That was an issue in 2017, and we are on a very positive trajectory. I get a brief every day from every one of the units on what the manning levels are.

The entire Navy is short a number of billets at sea, but the [forward-deployed naval forces in Japan] are manned to the level required. We are green across the board. That’s the manning piece.

The other is the circadian rhythm piece. I’m a submariner, and we went through this change in our watch bill a number of years ago. And look, life aboard a ship is hard, some days are harder than others. And it takes a while to make sure that everybody gets enough rest and is in the circadian rhythm. And you could be in it for five days and then get knocked out of it because something comes up during the time you are supposed to get sleep.

I think by and large the leadership on the ships know that crews need the proper amount of sleep, and they are focused on that. And I’m positive that at various times it will be disrupted.

What has been the feedback from the deckplates that you are hearing vis-a-vis the changes that have been put in place since the accidents?

There have been a lot of changes in the past year. And that can unnerve some people. My view is that all the changes we’ve made have been positive, and while there is some natural skepticism about how well this is going to be implemented, I think if you talked to the ships and their crews you’d find they are satisfied with us holding fast to the force-generation model — keeping the training sacrosanct, making sure they are getting the training they are supposed to get. I think if you talked to the waterfront, they’d tell you the same.

There seem to be an issue with maintenance throughput in Japan. How is that working now? For example, the destroyer Barry recently had an extended availability because of issues that arose during its ready-for-sea assessment.

I think maintenance is one of the key issues out there. I do believe there is a backlog of maintenance that we have been working through, and we’re not yet through that. And as a result, the [availabilities] are going longer. And when that happens, that causes perturbations through the entire force-generation model. And Barry is a good example of that. McCain, to be honest, is another good example. We expected it out at a certain point, but when they got in there they found additional work that needed to be done and extended it out.

I do think there are some improvements we need; whether its capacity or processes, I’m not precisely sure. But we need to get better at our [availabilities], those are the areas that historically I’ve seen and continue to see overruns. And we are still figuring out exactly what’s causing them.

The maintenance operation in Japan has been under some scrutiny in recent years, including a report by the U.S. Government Accountability Office prior to the accident. I’ve never heard anything bad about the quality of the workmanship in Japan, but the throughput is still down. Is there are program in place to get your arms around it?

The quality if very good, as you suggested. I have no complaints about the quality. It’s more about their ability to get them out in the time that has been allocated. I’m not sure if its capacity or process, but we have both [Naval Sea Systems Command] and [U.S. Pacific Fleet] taking a hard look at the [ship repair facility in Japan] to figure out what we can do to make the throughput what it should be.