Feb 22

Feb 22 Overhaul9: Navy Maintenance Problems3

Introduction

Navy ship depot maintenance has not been going well since 2014 according to Government Accountability Office (GAO) reports. I thought it would be interesting to use a few of the reports to understand (not solve, people have day jobs to do that) some of the difficulties shipyards (SY) face in supporting Navy ship material readiness. Unlike prior posts about the roles of ship crews in depot maintenance, I haven’t attempted to describe the perspective of people working in shipyards. I don’t have the data for that and can’t argue that what I learned from my Engineering Duty Officer experience is still relevant.

In this post, I’m going to describe success factors for depot maintenance, causes for depot maintenance delays that SYs can influence (if your solution does focus on those, you’re just dreaming), level of analysis issues with GAO assessments, and provide my take on what makes naval SYs different from other activities that execute large-scale industrial projects like construction. This post IS NOT a critique of GAO assessments of SY performance.

Factors Necessary for Ship Overhaul Success

In my two previous posts, I reviewed observations from GAO-20-257T NAVY MAINTENANCE-Persistent and Substantial Ship and Submarine Maintenance Delays (Dec 2019). The report and many others published 2014 to 2023 analyzed problems encountered delivering ships from depot maintenance (i.e., SYs) on time for the planned budget. The reports are consistent: ships deliver from maintenance late, costs are higher than expected, and maintenance backlogs (only computed when the GAO or Congress asks, the reasons for which are beyond the scope of my blog) have consistently increased.

By describing the causes of maintenance delays, the GAO indirectly identified the factors necessary for SY overhaul success. My list is:

sufficient (trained) workers (noted in GAO reports),
material, facilities (they don’t have to be modern, only sufficient, partially noted in GAO reports),
efficient maintenance processes (not included in GAO reports), and
maintenance planning that minimizes delays caused by new, unplanned work (see the graphic from GAO-20-257T, “Factors Affecting Maintenance Performance,” partially addressed in GAO reports).

The GAO reports don’t mention the maintenance processes used by NAVSEA (responsible for all ship maintenance), SYs, and type commanders. From my experience with Carrier Team One (possible future post), the processes used to plan and perform maintenance are just as important as having sufficient resources, which is why I put them in the list above. Bad processes seldom produce good results no matter how much money you throw at them. Navy ship maintenance processes include things like how Availability Work Packages (AWPs) are designed and negotiated, how work is coordinated during availabilities, how non-SY responsible work is planned and integrated into the availability schedule, how work control is managed, and how SYs and key constituencies negotiate risks and problems before availabilities start. Dysfunction or lack of compliance with the processes used to manage these things can cause delays, just don’t expect anyone to say that out loud. It is common for participants in processes to treat them as “just the way things are,” unless they work for Toyota. Toyota is known for testing, questioning, and improving its processes with a discipline that isn’t easy to match without decades (yes, decades) of practice. I’ll boldly claim that any organization that doesn’t regularly review and update its processes is headed for (or already in) trouble..

The GAO reports discuss new work, but don’t go much beyond “it causes delays.” One definition of new work is maintenance or testing that was not part of the AWP or SY schedule by the cutoff of work identification. New work has been a part of ship construction and maintenance since Joshua Davies and the original six frigates of the United States Navy. A key insight is it is seldom a “stuff happens” phenomenon. Without going into a lot more detail that is beyond the scope of this post, SY processes can mitigate much of the risk of new work delays.

Overhaul Delay Factors Under SY Control

In my previous post, I addressed the most important causes for ship overhaul delays identified by the GAO in their reports. The factors that SYs and NAVSEA can most readily influence are:

SY capacity (personnel and facilities),
material (ship overhaul requires lots of steel, cable, pipes, welding rods, and KimWipes)
SY performance, and
new work (to an extent, as noted above).

When I refer to SY capacity, I mean specifically the personnel and facilities in place at the start of the overhaul. For the foreseeable future, SYs have all the resources they are going to get. Hiring more personnel is controlled by Congress and training them takes years. SYs can only influence the bulk material needed for overhauls. Procuring spare parts for complex, high-performance weapons systems is not their responsibility. SY performance includes project planning, writing detailed repair instructions, worker efficiency and rates of re-work, SY industrial support processes, and resource management. GAO-20-588, Navy Shipyards: Actions Needed to Address the Main Factors Causing Maintenance Delays for Aircraft Carriers and Submarines (Aug 20, 2020) included comments on SY resource management.

GAO Levels of Analysis

I've noted that GAO analyses are limited by constraints affecting any group of investigators that haven’t spent decades working in SYs. It is a simplification to say that GAO analysts only know what people tell them, but it is still broadly true. Secondly, the analysis is limited by the data and subjective opinions of people in the organizations studied. Finally, the GAO assessed SY performance based on outcomes: cost and schedule. They suggested causal factors, but only at a high level like “new work impacts” delivery. This is a meso-level analysis of SY performance. Other levels of analysis are just as important for understanding problems with ship depot maintenance performance. People can debate where to draw the line in definitions of levels of analysis. My lines are:

Micro-level issues: wrench-turning (skills and experience of SY workers), training, internal SY process design and performance (sometimes they are not very efficient, HORRORS), maintenance process design that includes parties outside the SY, and how well ship crews support or don’t support the SY (discussed in a previous post).
Macro-level: Availability Work Package (AWP) design (how they’re created, negotiated, and approved, not what’s in them), ship designs that make maintenance difficult and costly, and “optimistic sustainment assumptions” by Navy Program Offices responsible for ship design and construction (GAO-20-257T and GAO-20-2, Navy Shipbuilding: Increasing Focus on Sustainment Early in the Acquisition Process Could Save Billions, Mar 2020). If Program Office personnel make optimistic sustainment assumptions that don’t come true (they never do), the effect will be that SYs need more resources to maintain ships. This isn’t insight. It’s just physics.

Both the micro- and macro-levels of analysis conceal “undiscussables” of Navy depot maintenance (Argyris, 2003). An undiscussable is something that people in an organization can’t or won’t talk about openly and the reasons for its undiscussability cannot be discussed either. The undiscussables of Navy ship lifecycle management (depot maintenance is just one part) include:

mismatches between the Congressional budget process and the work discovery (it doesn’t matter what the SY finds when they dive into a system for maintenance, no more money is coming from Congress),
management of SY capacity for future work (facilities, workforce size, and training),
ship design for sustainability (or not, see GAO-20-2), and
conflicts between maintenance and ship employment in pursuit of national priorities, sometimes maintenance is NOT the priority even though decisions not to follow maintenance processes adds costs and possible delays to subsequent maintenance (nobody remembers THAT when the ship finishes late).

The intent of this blog isn’t to criticize SY performance and processes, NAVSEA management of them, or take issue with the level of analysis chosen by the GAO to understand SY performance. The point IS to illustrate that Navy ship depot maintenance performance has lots of moving parts, many beyond the control of naval SYs or any single Navy organization. The GAO reports only tell part of the life cycle management story.

In the final analysis, omitting important parts of the “story” of Navy ship depot maintenance may not matter. All that NAVSEA, the SYs, and other interested parties can do is manage what they can control.

Naval Shipyards Are Different

There is a big difference between identifying problems with depot maintenance of Navy ships like late delivery, knowing how to manage all the things that have to go right to deliver on time, and having leverage to do anything about the most important causal factors like new work and ships being delivered from new construction with serious defects (GAO-20-2). In this sense, the GAO reports are like government improvement theater.

You can learn a lot about SYs from the GAO reports, but some important things are missing, which is where my blog comes in. One is that SYs have characteristics common to many large industrial organizations like car manufacturers, but are significantly different in three important ways:

Interactions, management, and coordination among multiple independent workforces (SY workers, private ship repair companies, private firms contracted by other Navy organizations, and ship’s crews). The involvement of multiple workforces isn’t unique to SYs, but their being independent of SY control IS. Can you imagine what car repairs would cost if you were allowed to help the mechanic?
Navy budgeting and work planning processes that are unlike anything in the private sector and are sometimes inflexible when unexpected things happen or national priorities change.
Repair work sounds repeatable (tank refurbishment, valve repair, switchboard re-wiring, etc.), but isn’t really. No two ships are exactly the same, valves have different materials, configurations, and designs to choose just one item.
Goal competition between maximizing the use of ships for national security priorities (operations), super sexy warship capabilities (design), human resources (people are expensive), and compliance with NAVSEA processes created for maintenance on vessels intended to go into harm’s way (conservative and expensive).

Conclusion

There are no solutions in ship lifecycle management, only tradeoffs among goals like depot maintenance predictability, SY capacity, warfighting capability, crew manning, operations, and acquisition. Armchair Chiefs of Naval Operations or people who have “fantasy plans” for improvement describe great clean sheet approaches for ship design, naval SY processes, national strategy, weapons system capabilities, or procurement and contracting regulations. Intriguing as the problem diagnoses and magic-wand-enabled solutions can be, they don’t work for complex processes like Navy ship depot maintenance because they involve different Navy commands (“stakeholders” is the cliché), maintenance activities, Congress, and defense contractors. These groups aren’t always aligned.

The reality of SY manning and facilities limitations is that even the GAO concluded that they are not going to change significantly in the next five years (GAO-23-106673). Ships aren’t going to get any less complex. Demands for using them won’t decrease and won’t be affected by potential impacts to future maintenance. In my next post, I will discuss what I think “facing facts” means for Navy ship depot maintenance because I can’t find a magic wand!