The Navy has had persistently poor shipbuilding outcomes, schedule delays, and cost overruns over the past decade because policy and processes allow the service to deviate from best practices, a new Government Accountability Office (GAO) report says.

Reviewing and updating a decade’s worth of reports and recommendations, the GAO found that the Navy has fallen 50 ships short, gone $11 billion over budget, experienced many schedule delays, and delivered ships of poorer quality than expected.

The future USS Gerald R. Ford (CVN 78) arrives at Naval Station Norfolk in Virginia on April 14, 2017, after returning from builder's sea trials. (Photo by U.S. Navy)
The future USS Gerald R. Ford (CVN 78) arrives at Naval Station Norfolk in Virginia on April 14, 2017, after returning from builder’s sea trials. (Photo by U.S. Navy)

GAO used a starting point of the Navy’s 2007 Long-Range Shipbuilding Plan and found significant problems compared to what the service had planned for. The Navy had envisioned 330 ships by 2018, but only has 280 currently while at the same time getting $24 billion more in funding than it planned for in the period.

The report noted ships routinely cost more and take longer to build than expected. GAO highlighted that lead ships, which are expected to experience some cost growth, experienced $8 billion in cost growth plus years of schedule delays in the past decade. It also noted that “many of these ships were provided to the fleet with less capability and poorer quality than expected.”

In the past decade lead ships cost $8 billion more to build than initially budgeted for in the 11 more recently delivered lead ships. Indeed, three lead ships had costs grow over 80 percent higher than initial budgets: the first Freedom-variant (LCS-1) and Independence-variant (LCS-2) Littoral Combat Ships (LCS-1) at 150 percent and 154 percent, respectively, and the San Antonio-class amphibious transport dock (LPD-17) at 80 percent.

In contrast, the initial Spearhead-class expeditionary fast transport (T-EPF-1) and lead expeditionary transfer dock (T-ESD-1) came through at one and three percent under budget, respectively.

GAO also found many follow-on ships, while experiencing lest cost growth than lead ships, “budgets for follow-on ships have generally still been higher than the Navy’s initial planned cost per ship for the class.”

For example, follow-on DDG–51 Flight IIA ships were almost nine percent over budget, LCSs were almost 13 percent over budget, LPD-17s were 2.6 percent over budget, and Virginia-class SSN-774 submarines were almost 40 percent over budget.

Beyond growth in average ship costs, the Navy has also received $3 billion in subsequent funding to finish all follow-on ships as well.

GAO also found regular shortfalls in schedule expectations.

“Schedule delays are common in Navy shipbuilding programs. All 8 of the lead ships we have reviewed were provided to the fleet behind schedule, and more than half of these ships were delayed by more than 2 years.”

For example, the lead Gerald R. Ford aircraft carrier CVN-78 had its schedule slip by 25 months, the first Zumwalt-class destroyer DDG-1000 slipped 72 months, LCS-1 slipped 18 months, the first America-class amphibious assault ship LHA-6 slipped 18 months, LPD-17 slipped 54 months, and SSN-775 was delayed 29 months.

Follow-on ships also experienced schedule delays, most notably, “deliveries of almost all LCS under contract (LCS 5-28) have been delayed by several months, and, in some cases, a year or longer” while LPD-17 ships were delivered 15-20 months behind schedule.

These kinds of outcomes reduced the Navy’s buying power, contributing to the shortfalls in the Navy’s 2017 shipbuilding plan.

GAO found a main cause is “that the Navy does not consistently follow shipbuilding best practices, which emphasize the attainment of sufficient levels of knowledge at key milestones and could improve acquisition outcomes.”

“We have found that the Navy routinely accepts delivery of ships with large numbers of uncorrected deficiencies, including starred deficiencies, which are the most serious deficiencies for operational or safety reasons,” the report said.

That conflicts with Navy policy that states vessels should be fully mission capable with all contractual responsibilities finished before delivery aside from requirements like ship outfitting.

The future USS Portland (LPD-27), sails in the Gulf of Mexico during acceptance sea trials. (Photo: Huntington Ingalls Industries)
The future USS Portland (LPD-27), sails in the Gulf of Mexico during acceptance sea trials. (Photo: Huntington Ingalls Industries)

This started to improve after 2009 when the Navy began taking steps to improve oversight of ship construction following several cases of poor ship quality.

“Since then, the Navy has generally reduced the number of uncorrected deficiencies at the time of ship delivery, particularly for follow-on ships,” GAO said.

However, the office found the Navy is still accepting ships with uncorrected starred deficiencies. In 2017 the GAO found “90 percent of acceptance trial starred deficiencies were not corrected prior to delivery for the eight ships we reviewed.”

GAO also pointed out many ships do not meet performance expectations. Only half of the six ship classes that underwent operational testing in the past decade passed it on the first attempt: SSN-774, T-AKE, and LHA-6.

Reliability problems may not result in a ship class failing operational testing, but unreliable systems provided to the fleet without correction can consume limited resources, inhibit performance, and add to sailors’ workload, the GAO said.

The GAO said there is a disparity between resources planned to execute programs and the capabilities the Navy looks to acquire.

“This imbalance forms during the pursuit to fund lead ship construction, when competitive pressures to get funding for the program are high and many aspects of the program remain unknown. During this process, the Navy often initiates shipbuilding programs with weak business cases that over-promise the capability the Navy can deliver within the planned costs and schedule.”

Then when ship construction starts and the initial business cases erode, shipbuilding comes under pressure to control cost and schedule growth, usually by changing the planned quality and performance goals. By the time the pressures become real, multiple ships are usually under construction and the changes cause disruptions throughout the ship class.

“Over time, this approach reduces the Navy’s buying power and the likelihood of achieving fleet goals,” the report said.

Rather than gradually building knowledge over time and sequentially moving through the shipbuilding phases of technology development, design, and construction, the Navy often experiences overlap.

“Concurrency in shipbuilding programs can frequently result in the opposite of its intended results. Instead of recovering schedule losses, concurrency typically results in cost growth and schedule delays. If technologies are not mature before the ship is designed, there will likely be design changes later in the shipbuilding process when the technical requirements become more well-defined.”

Then if design changes are made after construction is already started, the shipbuilder will have to complete out-of-sequence construction and redo completed work to implement the changes.

Indeed, the GAO found that most ships it reviewed started construction before technology development and ship design were finished.

“We have found that the programs with the greatest amount of overlap between shipbuilding phases often have some of the highest cost and schedule growth, as well as quality and performance issues,” it said.

The report cites Ford-class carriers as a case study in which “a weak business case for the Ford-class aircraft carrier led to poor outcomes.”

Back in August 2007, the GAO found the business case for CVN-78 “was predicated on unrealistic cost and schedule estimates that did not sufficiently account for risks associated with the development of the Ford class’s 13 critical technologies and construction of the lead carrier.”

USS Washington (SSN 787), a Virginia-class submarine, completing initial sea trials. Photo: Ashley Major/Huntington Ingalls Industries.
USS Washington (SSN 787), a Virginia-class submarine, completing initial sea trials. Photo: Ashley Major/Huntington Ingalls Industries.

Although lead ships usually have challenging technology development and construction issues and require more funding and labor hours than follow-on ships, the Navy estimated CVN-78 would require fewer labor hours than the last two previous Nimitz-class carrier.

This decision was exacerbated by the service’s decision to start construction while still developing important technologies like aircraft launch and recovery systems, radar, and weapons elevators. This caused “design changes and construction disruptions that led to cost and schedule growth,” GAO said.

In the end, the Navy took delivery of CVN-78 in May 2017 two years late and $2 billion over the initial budget. The ship will not be ready to deploy until 2022 because major development, construction, and testing are not finished.

“Despite more than 15 years of development, reliability and performance shortfalls with several of the carrier’s key technologies persist—such as with the aircraft landing system,” GAO said.

GAO said the Navy has started taking some actions to improve knowledge and reduce risk before key milestones, but it may take years before it sees the outcomes of changes in long shipbuilding timelines. Moreover, reforms like reducing technology risk and increasing design stability cannot be fully applied to current ships and will only be offered to new ship classes, set to start construction in the early 2020s.