Arsenal Ship Lessons Learned Report
1.0 Executive Summary
1.1 IntroductionThis report captures many of the lessons learned in executing the first two phases of the Arsenal Ship program, a period of about 22 months. It focuses on acquisition process-oriented lessons, rather than individual consortia technology initiatives which are proprietary in nature. This report was written by the Arsenal Ship Joint Program Office (biographies at Tab M), and the views expressed do not necessarily reflect DARPA or Navy concurrence.
The first two phases of the Arsenal Ship Program provided an excellent return on the Navy's and DARPA's investment. These two phases successfully demonstrated that industry, involved early in the ship design process, could develop an optimum mix of performance capabilities that could be accommodated within affordability constraints; successfully demonstrated teaming between combat system integrators and shipyards; and introduced innovative concepts in reduced manning, automated damage control, topside integration, and modular design.
1.2 BackgroundThe basic requirement for the Arsenal Ship, established in a joint Navy DARPA Memorandum on March 18, 1996 (Tab A), was to satisfy joint naval expeditionary force warfighting requirements in regional conflicts by providing the theater commander with massive firepower, long range strike, and flexible targeting and possible theater defense through the availability of hundreds of vertical launch system (VLS) cells. To meet this warfare requirement affordably, the Arsenal Ship concept and design was to be straightforward and simple. Detailed requirements and concept of operations were defined in separate documentation (nine pages total), however, key elements for the Arsenal Ship included:
In addition, cost was viewed as an independent variable, and early industry involvement with the development of a cooperative industry-government team was viewed as key to achieve Arsenal Ship goals. To minimize cost, off-the-shelf systems were to be used exclusively. Any development of new systems required the approval of ASN (RD&A). The cost of acquiring the first ship was not to exceed $541 million including the cost of concept development and competition. These funds were to be provided jointly by the Navy and DARPA with contributions of $371 million and $170 million respectively.
The non-ACAT Arsenal Ship demonstration program was created to evaluate sea-based massed precision firepower, while minimizing the risks in acquisition of approximately six ships. To ensure that the program remained affordable, a firm acquisition cost threshold for the production ships was established (Unit Sailaway Price $450M goal, $550M cap in FY96 $). A corresponding Life Cycle Cost threshold was also established (Operating and Support annual cost of $13.7M/ship [roughly a third of DDG-51]. This program was conducted using DARPA's Section 845 Agreements Authority so as to allow industry wide latitude in satisfying the Navy's requirements within this threshold. Agreements were structured to allow tradeoffs between cost and performance. Program success was to have been judged by the extent to which the Arsenal Ship met operational requirements.
A second purpose for this demonstration program was to accelerate the Navy's ongoing acquisition reform activities focused on buying improved ships at a lower cost. To this end, the joint program was to focus on exploiting DARPA's culture and experience in prototyping system programs. The Navy and DARPA anticipated the production Arsenal Ship contracts would serve as a model for future streamlining.
This joint Navy/DARPA demonstration program was conducted under DARPA lead, as articulated in a joint 28 May 1996 memorandum (Tab B), with an envisioned transition of leadership to the Navy in the testing and production stages of the program. The program was managed by a joint Navy/DARPA program office with the Program Manager reporting to DARPA. A small program office was mandated. DARPA, Naval Sea Systems Command (NAVSEA), and the Office of Naval Research (ONR) each provided two billets. It was expected that the program office would grow to a maximum of three billets each as the program grew to maturity. The Navy developed a concept of operations (CONOPS) (Tab C), and a Ship Capabilities Document (SCD) (Tab D), which the program office used to guide the Phase I and II trade studies conducted by industry. The Program Manager developed a program plan including major decision milestones, and a DARPA to Navy program transition plan.
Organizationally, the Arsenal Ship Program Manager reported to two committees for guidance and direction: the Steering Committee and the Executive Committee.
The Steering Committee included:
Deputy Assistant Secretary of the Navy (DASN, Ships)
Assistant Director, TTO for Maritime Programs - DARPA
Director, Surface Warfare Plans/Programs/Requirements Branch
Office of Naval Research (ONR33)
Assistant Secretary of the Navy (RD&A)
Director of Surface Warfare (N86)
Chief of Naval Research
Five Teams were awarded Phase I agreements on July 11, 1996, under the 10 USC 845 Other Agreements Authority, with the potential to execute through Phase VI of the program. This Other Agreements Authority excepted the program from the Federal Acquisition Regulations (FAR) and any military specifications and standards. The scope of Section 845 Prototyping Authority is explored fully in a DARPA General Counsel October 24, 1996, memorandum (Tab E). USD(A&T) implementation of 845 Other Agreements Authority within the Services (Section 804 of the FY97 Authorization Act) is included at Tab F.
Three Teams were awarded Phase II agreements for Functional Design effort on January 9, 1997. In April 1997, the original Arsenal Ship Demonstrator concept was expanded by the Navy and DARPA to include risk reduction efforts for the Surface Combatant of the 21st Century (SC-21) (Tab G) and the Demonstrator's name was changed to the Maritime Fire Support Demonstrator (MFSD) (Tab H) to reflect this expansion.
The FY98 Appropriation Act appropriated $35 million of the requested $150.2 million for Maritime Fire Support Demonstrator. On 24 October 1997, the Secretary of the Navy reluctantly determined that continuation of MFSD into Phase III and further phases was unachievable (Tab I). Phase II activities were completed and the Arsenal Ship Joint Program Office closed operation on 31 December 1997 (Tab J).
1.3 Lessons Learned1.3.1 Acquisition streamlining works. The process being followed by Arsenal Ship demonstrated a 50% reduction in acquisition time for the design portion of the ship compared to the traditional design approach. The Industry Teams were prepared to complete the detail design and build the ship in 33 months, again about half the normal time. This was primarily enabled by using an industry led acquisition operating under Section 845 authority, with industry having full trade space and responsibility for the design.
1.3.2 Price As Established (PAE) spurs innovation and drives down acquisition cost but increases risk. Price as established (PAE) was the approach taken throughout the Arsenal Ship Program. Price here means industry cost to manufacture and a reasonable profit (or return of investment to the company). A price goal was established early in the program. All designs were monitored by the contractor teams to ensure an affordable item at a price including all non-recurring and recurring costs to manufacture and a reasonable return on corporate investments or in other words, profit. All other aspects of the government desired capabilities were tradable against the price goal. This is different from Cost As an Independent Variable (CAIV). Cost as independent variable (CAIV) assumes two things; first, the government is monitoring and controlling the trade decisions to ensure affordability, and second, cost is only one of several factors to be traded. Requirements creep is most certain. CAIV in present government literature means a government program manager yardstick to consider trades against requirements.
After several months, all competitive teams were to fix their own price against their design or PAE. During later phases more trades could result. The contractor team should have responsibility for all trade decisions and be encouraged to use the trade space within the government's desired capabilities to ensure an acceptable and capable product that meets PAE. PAE focused the design trades on mission essential performance, without requiring the use of legacy systems as GFE (Government Furnished Equipment). This resulted in designs that were engineered at a total system level, highly integrated, commercially-based, and low cost, while at the same time introducing new systems with some attendant risk. Industry Teams were prepared to meet the goal of $450M for the average production ship cost, about 1/3 less than early Navy estimates for the production ship. Risk reduction programs were key to achieving the PAE goals and worked effectively. Under the Arsenal Ship acquisition strategy, substantial elements of the risk were transferred from Government responsibility to Industry. In short, competing teams concluded "zero risk is unaffordable."
1.3.3 Low manning is readily achievable. Manning is the largest single factor in life cycle cost, and accounts for roughly 40% of the annual operating and support costs of Navy ships. Reduced manning is the key to lower life cycle costs. The enablers to achieve lower manning among the Industry Teams included: commercial ship operating and maintenance practices, insertion of COTS technologies such as integrated bridge system; nested, remote sensors; and high levels of controls and C4I integration. The Arsenal Ship goal of less than 50 crew was easily achieved by the Industry Teams (who averaged 22 crew). Preliminary Navy estimates for Arsenal Ship were 269 crew.
1.3.4 An industry design competition could be more meaningful than a Government AOA (Assessment of Alternatives). Industry Teams produced an array of design solutions that were achievable and affordable, with better cost information than is possible within the Government's data base. Since the designs were optimized for each Team's production capabilities and facilities, industry design alternatives were available at lower cost than Government designs, and had been measured against producibility metrics, so designs were at a much higher state of maturity.
1.3.5 Industry is fully capable of designing and developing complex Navy ships. The majority of design and production skills needed to produce Navy ships was assembled as a natural part of the teaming of the full service consortia. Industry Teams were able to find needed technical expertise for almost all design areas in the commercial marketplace. In selected cases (principally survivability, and weapons effects), some Government R&D Center expertise was obtained by industry under individual contract. During the downselect process, losing Team members became available to join the winners to develop even stronger Teams. This was encouraged in the Agreement between the Government and the Industry Team.
1.3.6 Significant cost savings in development and acquisition programs can only be achieved through program stability. Industry's willingness to invest its time, talent and resources to compete for 845 type agreements is strongly influenced by the clarity and stability of the program, as well as the ability to realize return on investment. The Arsenal Ship Joint Program Office left the CONOPS and key elements of the Arsenal Ship unchanged through the first two phases; this stimulated about a $5 million per team investment (Consortia IR&D) in Phase I, and approximately $15 million per team investment in Phase II. The potential for achieving a reasonable return on investment for production Arsenal Ships, as well as future production SC-21 vessels rounded out Industry's motivation for their investment strategy.
1.3.7 Technology is already available for breakthrough performance. Technologies still "under study" by the Government were readily incorporated in the Arsenal Ship designs based on COTS products. Areas of particular strength include: reduced manning; automation; information systems; communications/connectivity; propulsion machinery; fire fighting; maintenance and logistics. Effective passive survivability and signature reduction technologies derived from previous Government programs are also available in the Marine and Aerospace industry. Combined, these resulted in improved performance with major savings for both acquisition and service life costs at little technical risk.
1.3.8 Minimal Government direction is a key factor to success. In a typical Government acquisition program, extensive specifications, cost and schedule requirements, and oversight (DoDI 5000.2, as interpreted by program offices and oversight staffs) can result in an overly constrained environment and unachievable objectives. Further, large Government program offices provide multiple opportunities for redirection, levying of additional requirements, confusion and delay in administering acquisitions. Arsenal Ship, with a minimal set of technical objectives (9 pages of goals with no thresholds), coupled with an office size of six Government employees, kept industry's trade space open, communications direct, and delays to a minimum. Significant trust and mutual respect was a direct consequence of the more open and coherent dialog with Industry.
1.3.9 Adequate time is needed for Industry Team formation and growth. Much energy was spent by industry in the first two months of Phase I sorting out Team membership and Team relationships. At the start of Phase II, the average Team size increased from about 50 to over 200 people in a short time frame, which caused different design maturation rates among the Teams. The phases of the program were structured to allow seamless transition from one phase to the next. Industry did not take full advantage of the downselect decision period to position themselves for the next phase. Had they done so, the Phase transition would have been much smoother. Without this, an additional 2 months for Phase II would have permitted all three consortia to fully mature their functional design.
1.3.10 Competitive design solutions are a "package deal". All aspects of the Industry Team's competing designs were not equal in performance, but each Team met the PAE goal without containing any fatal design flaws. Achieving a notional "best possible ship" by having the Government attempt to integrate various aspects of each Industry design would have resulted in (a) holdback of the most innovative ideas from competing Teams (b) a Government directed solution, undermining the industry-based design responsibility and (c) a program that would not meet its PAE goal. The Government must understand (and participate in the underlying trades which lead to the winning design), and then accept the solution as the best compromise that can be achieved within the PAE constraint.
1.3.11 Section 845 permits "fly before buy" for Naval Ships with no lost time to full production. The Defense Science Board's 1996 Task Force on Defense Acquisition Reform (Phase III) highlighted the time to field the average major weapon system at 16 to 18 years. Surface Navy ships have been no exception. Traditional surface combatant acquisition programs require five years to complete concept, functional and contract design, after a five to seven year research and development cycle. Following award of lead ship, production of the first vessel averages five years. Two years after lead ship award, the follow ship contract is let, when the lead ship has barely started construction, committing the Navy to procurement of many ships before the first one is even delivered.
Using Section 845 authority, and involving industry at the inception of the program, cut the development cycle in half. The Arsenal Ship demonstrator would have been available for testing after just four years of design and production. The production decision could then be based on actual test data, allowing a ship to be acquired in a "fly before buy" fashion, reducing cycle time and accelerating technology insertion into the fleet.
1.3.12 R&D Funding must be properly balanced with production costs. Industry's limit of $389M for development of the demonstrator ship was extremely challenging as compared to the average production cost goal of $450M. Industry pushed some software development to the production ship to reduce demonstrator ship non-recurring costs, with attendant reduction in demonstrator capabilities and increased production risks.
1.3.13 Industry lead for acquisition is a permanent "fork in the road". Industry's freedom to balance the PAE tradespace equation necessarily includes their ability not to choose Government developed systems. This has implications for the Government's R&D investment strategy, the infrastructure currently in place to support subsystem development (Participating Managers [PARMS]), and equipment configuration control (outsourced and privatized to industry or centralized with the Government). The net effect could be to lower total Life Cycle Costs by attacking infrastructure as well as taking advantage of the commercial marketplace.
|Acquisition Reform Office
Ms. Carol Szot, Webmaster
Send all comments/questions to: email@example.com
|Last revision: 10 March 1998|