The sinking of the Israeli destroyer Eilat by surface-to-surface missiles in 1967 led to an urgent requirement to update both the early warning and immediate response capabilities of the U. S. surface fleet. Theoretically, the best method of providing early warning is through electronic warfare (EW). However, in the years following 1967, several efforts which had gone as far as the prototype stage had been priced far beyond the Navy’s ability to pay. By 1972, the Navy was not much closer to replacing its mid-1950s vintage surface EW equipment than when the Eilat went down. Somehow, performance requirements had to be brought in line with the fact that a finite—if slightly flexible—amount of dollars existed for the development and procurement of the desired number of pieces of equipment. In May 1972, there began a new program to develop a modular family of surface electronic warfare systems, the prices of which could vary from $500,000 to $3 million.
Origins of Design to Cost: A new procurement philosophy had originated in 1970 when Deputy Secretary of Defense David Packard concluded, “. . . overruns are the end product of our mistakes rather than the key issue to be addressed.”[1] Dr. John Foster, then Director of Defense Research and Engineering, was the first to try to come to grips with a new procedure to solve the cost overrun problem. He explained:
“We must design to a price. We must revamp the design philosophy in the Department of Defense and in defense industries. We shall not in the future indulge the present syndrome of incorporating into every system the most advanced technology as soon as it seems to be available or merely because it is advanced.”[2]
But it was two years before Dr. Foster was able to outline the new “design-to-a-price” technique; the procedures he set forth were ordered into effect wherever possible. The Navy’s requirement for a piece of surface EW gear was among the first programs to have design-to-cost controls applied from the beginning. Consequently, that program can be examined as a case study by which to evaluate the potential effectiveness of the new system.
Previous Efforts to Control Costs: To understand better how the design-to-a-cost philosophy has fared in its application to one particular hardware development program, it is instructive to review why previous efforts aimed at reforming the weapons acquisition process have fallen short.
Before the mid-1950s, little attempt was made by the separate military services to coordinate their development planning or their acquisition policies. “Cost-plus” contracts were used almost exclusively.[3] By the mid-1950s, inflation and the increased sophistication of weapons technology had begun to drive costs up sharply, and the need to impose some sort of controls was widely recognized. Early efforts focused on “production prototype philosophy,” which meant making more incremental efforts at development but with added stress on equipment reliability and cost control. Problems continued to exist, however, because of a lack of appreciation of the growing complexity of modern weapon systems, the lack of logistic support for new weapon systems, and continued inadequate reliability of the equipments themselves.
In 1961, Secretary of Defense Robert McNamara came into his new post determined to get control. Some of his better known changes were the new Planning, Programming, and Budgeting System which included a five-year planning cycle, the organization of budget requests by program category rather than by each service’s “shopping list,” rigorous quantitative analysis of service budget requests, and the use of paper design competitions to determine source selections. Other reforms of the McNamara era were the use of fixed-price contracts to control costs and, later, the adoption of “total package procurement” in an effort to purchase a new weapon system and all its related support more efficiently.
McNamara succeeded only in revolutionizing the methods by which the Department of Defense was mismanaged. Symptomatic problems such as cost overruns, contract claims, “buy-ins,” “bail-outs,” and defective systems were addressed on a piecemeal basis.[4] Little was understood about the process itself, or about the key role of the individuals in the process. Before long, the services learned the new “rules” of the game and were able to circumvent them or turn them to their own purposes. Contractors also learned the rules and insisted that redetermination and escalation clauses be written into fixed-price contracts. These “escape clauses” later produced the series of horror stories which peaked at the beginning of the 1970s. “Fixed-price” contracts, as it turned out, failed because of unenforceable controls and unrealistic prices. In the end, they controlled neither prices nor costs.
Total package procurement included development, production, and spare-part support all in the same fixed-price-plus-incentive contract. The advantages expected were greater efficiency in procurement by sole-source contracting and long-term commitments, the inhibition of “buy-ins,” and greater program reliability by shifting the major burden of risk and management to the contractor. Total package procurement was thought to be the great hope of the future in defense contracting, but the results included the cost overruns of the F-111 and C-5A aircraft and the Spruance (DD-963)-class destroyer. In 1972, total package procurement was officially abolished by the Deputy Secretary of Defense.
Design to Cost: The new system can be understood as a series of related corrective measures for the failings of total package procurement. Comparison of the two is depicted in Table 1. Included within the new concept are:
► Increasing the size of individual equipment purchases whenever possible to reduce unit costs
► Providing greater standardization in system design by reducing each service’s proclivity for optimizing everything around individual platforms and missions and by employing modular building block construction whenever possible
► Providing controls to prevent, on the one hand, the service from making incremental changes in specifications once the equipment is well along in development, while on the other hand preventing the contractor from attempting to add every new state-of-the-art advance to the equipment whether it is vital to the operational requirements or not
► Considering life-cycle costs in all analyses rather than only front-end costs such as development and procurement
► Requiring the contractor to assume a larger role in maintaining the equipment after purchase is made, along the lines of a contractor warranty
► Establishing better program management throughout development by providing increased program visibility, direct access of the program manager to high DoD levels, and retention of key personnel throughout a system’s development.
Table 1
Total Package Procurement versus Design-to-a-Cost
Area | Total Package Procurement (Failure) | Design to Cost (Solution) |
Quantity | Single early decision for specified numbers | (1) Experience buildup before quantity decision (2) Trade-offs and feasibility studies |
Specifications | Set by service in extensive detail | (1) Set general requirements; let contractor design to both cost and capability. (2) Maintain early flexibility |
Contract | Fixed-price throughout | (1) Cost plus incentive for development (2) Fixed-price for production run |
Source Selection | Paper design competition | (1) Prototype development (“fly-before-buy”) (2) Prototype competition (“fly-off”) |
The Current Weapons Development Process: All major defense acquisition programs (defined as any program in which development costs exceed $50 million, production costs exceed $200 million, or the Secretary of Defense so designates) are required to proceed through a set of hurdles: concept formulation, validation, and full-scale production. Few programs flow smoothly, and steps can be skipped or may have to be repeated several times. Above all, the formal development process is inseparable from the political nature of the arena in which it takes place.
Prior to proceeding from one phase to the next, a program must be reviewed and approved by a Defense Systems Acquisition Review Council (dsarc).[5] In general, programs are taken to dsarc when they need authority to spend more money. Thus, the dsarc is able to advise the Secretary of Defense on the management of defense programs at their critical decision points. In order to coordinate efforts and recognize any potential problems, the Navy runs its most critical programs through its own reviews.
Organization for the Program: For any given project the specific organization which develops—and the personal preferences and biases of the individuals involved—often goes a long way in determining the success of the program and the specific characteristics of the end product. Furthermore, the extent of involvement of each individual will change as the program moves along, based for one thing on professional and organizational interests and, more importantly, on who is paying for the project at the time.
Normally there are several key participants in the development of a major program. The “platform sponsor,” in this case the Deputy cno (Surface Warfare) (OP-03), is responsible for writing the original operational requirement, supervising the funds used for production and procurement, and representing the eventual user-branch (air, surface, or submarine). Since the platform sponsor will also eventually be the “user,” he is the one participant who has an interest in following the program through all of its phases, from design to installation.
The mission sponsor will be a second important Participant in most programs since he is charged with coordinating warfare efforts across parochial branch-of-service lines. In this case it was OP-095 (Antisubmarine Warfare and Tactical Electronic Warfare Programs), but as it happened, that organization played a relatively minor role because its reponsibilities were in flux throughout the period of the EW program. Another major participant is the program manager, who is charged with most of the day-to-day administration and supervision. Naval Electronics Systems Command (NavElex) was the program manager and gave the program office (PME-107) direct access to report at the top rather than through several tiers of internal hierarchy.
Since all costs up to the awarding of the production contract come out of the Navy’s research and development fund, the Director of Navy Research, Development, Test & Evaluation (OP-098) is sure to have some amount of say in how those funds will be used. OP-098’s normal role includes smoothing the rough spots for programs in development, levying political realities onto the platform sponsor’s requirements, and serving as a technical staff for operators who may have a feel for what it is they need but less appreciation of how it should work. Because of the newness of the design-to-cost concept, OP-098 turned out to be the leading spokesman for OpNav in defining, directing, and driving the EW program until full-scale production was approved.
Commander Operational Test and Evaluation Force (ComOpTEvFor) will often be a fifth important participant. OpTEvFor is responsible for supervising the operational testing of all new equipment and making recommendations to the cno as to its further development or adoption for service use. In the EW program, ComOpTEvFor arranged all services and schedules necessary for both operational (OpEval) and technical evaluation (TechEval), kept contractors happy regarding the at-sea test schedule, and, of course, evaluated test results during the OpEval. Figure 3 summarizes the Navy organization for the surface EW program.
Table 2
The Defense Acquisition Process
Phase | Concept Formulation | Validation (also known as “Contract Definition”) | Full-Scale Production |
---|---|---|---|
Associated DSARC | DSARC I | DSARC II | DSARC III |
objective(s) | • Develop cost criteria • Define performance criteria | • Design development • Fix production/cost goals | • Final approval • Resolve contract problems |
Preliminary Activities | • OR (Operational Requirement) set • Performance plan set • Cost estimates established • Delivery schedules prepared • Logistics plan developed • Test plan arranged | • Develop technical approaches • Prepare RFP • Proposals submitted • Testing • Contractor recommended | • Contract awarded • Production monitoring |
Subsequent Activities | • SecDef approval • Obtain congressional funding | • SecDef decides source selection | • Contractor delivery |
Key Actor (Navy) | OpNav | NavMat | NavMat (and contractor) |
The Surface EW Project, Initial Selection: Following the May 1972 CNO order which initiated the program, the Naval Material Command (NavMat)—in conjunction with OpNav—started drawing up plans. In addition to the basic design-to-cost guidelines outlined by Dr. Foster, some additional ground rules were established:
► Contractors were to be given freedom to design the equipment to satisfy the objective rather than rigid specifications.
► There would be competitive development between at least two contractors which would include a side-by-side testing to provide an incentive both for performance and for controlling assets.
► Off-the-shelf technology was to be used rather than designing new systems which pressed the state of the art.
► The equipments, and the program as a whole, were to be low risk. (Something had to be produced that would work, for the Navy was not going to experience yet another EW program that was terminated with nothing to show for it.)
Figure 3
In January 1973, the Navy asked six of the many bidding companies to develop their preliminary proposals. The survivors were Westinghouse, rca, Sanders, ail, Hughes Aircraft, and Raytheon. NavMat analyzed the proposals of these six companies and in October, following a four-month evaluation period and a month of dsarc hearings, contracts were let to Hughes Aircraft and Raytheon to build prototypes, SLQ-31 and SLQ-32 respectively, based on what seemed the most effective design proposals. Shipboard testing was scheduled to begin in late 1975.[6] It was determined that a series of three modular suites would be built; that the costs would be set at $300,000, $500,000, and $1.4 million respectively; and that “Suite-3” would be built as the competitive prototype.
The new EW suite would be designed to automatically review and match incoming signals to its own library and provide the operator with a complete data readout, visual display, and recommended countermeasures. The system would employ a modular building-block concept wherein different types of ships would have the same basic receiver capability, but some would have broader response capabilities depending upon the particular suite installed.[7]
Conduct of the Testing: The cruiser Leahy (CG-16) was designated as the project test ship for both of the competing EW suites. This was apparently the result of a compromise between fleet commanders. A cruiser was large enough to test the two systems side by side. This saved the use of one ship, because the original plan called for one ship to carry each competitor’s equipment. The Leahy was available because she was scheduled to transfer from the Atlantic Fleet to the Pacific and hence was not yet slated for any Particular tasks by her new fleet commander. Unfortunately, installation of the equipments could not begin until her return, in late October 1975, from a Mediterranean deployment.
A problem encountered once the testing began was that “ships sure don’t resemble backyard test lots,” as one company’s technician put it. Equipments which had worked perfectly in controlled
environments seemed hardly to work at all in their new surroundings. The dense electromagnetic field, the effects of water surface on sensors, and the perpetual ship motion all took their toll on the equipments in the early stages of the testing, and initial results were disappointing. Little improvement was noted after almost two months of testing, and in April 1977 the decision was made to halt all at-sea testing for two months while everyone fell back and regrouped. When testing resumed in July, dramatic improvements were noted in both equipments, and the tests progressed rapidly through to completion.
Bureaucratic Problems: During the period when problems were being encountered, the two contractors employed quite different tacks. While at the working level one contractor admitted having problems and suggested a delay of the tests, the other—whose equipment seemed to many to be operating only marginally better—sounded a more optimistic note and hinted that the Navy could declare the other contractor no longer competitive.
In view of the problems experienced in the early tests, the delay now seems to have been an obvious choice. But there were several important reasons why it might not have been in the best interests of the program:
► Since the contract specified firm schedules, the Navy could put itself in a legally untenable position since one contractor was calling for a continuation of the tests.
► Allowing contractors two more months to iron out problems involved the risk that they would use the time to add their latest “improvements” rather than fixing the problems which already existed.
► The Leahy was scheduled to start a major overhaul late in 1976. Extensive planning and preparation were necessary. The fleet commander was determined that she would start the overhaul on time, even if it meant the EW equipments had to be moved and tested on board another ship.
► In addition to the potential costs of having to move the equipments, time itself was expensive since the contractors were being paid on a monthly basis. NavElex might be forced to go back to CNO for the additional money, thus publicizing the fact that the program was not doing so well as preliminary reports had suggested.
This would mean not only embarrassment but also that the program manager would find himself in direct competition with other program managers who were like wolves at the door in always looking for more funds. If such funds were not readily forthcoming, the whole program might be cancelled, curtailed, or extended indefinitely.
► Finally, since it was already late in the fiscal year, it was a particularly undesirable time to have to admit to problems. Since congressional appropriations are made annually, funds have to be obligated during a given fiscal year or they are “lost.” Consequently, the “wolves” are always particularly alert late in the fiscal year to pick up the scent of any project which might be in trouble and which, as a result, might have some unused money lying about. That, in turn, could mean deferral, or even indefinite suspension of the project until more money is authorized.
Nevertheless, all these problems paled against the overriding concern over the possible loss of the competitive nature of the tests. If the field of contractors were narrowed to one, that contractor would be in the driver’s seat regarding costs, and the main controlling tool for the design-to-cost idea would be lost. Consequently, despite all the potential problems, the decision to delay the tests until both contractors could resume on a progressive and competitive basis turned out to be a sound one.
Attitudinal Problems: Several attitudinal problems observed during the tests are worthy of note since they tended to affect adversely the fleet’s ultimate receptiveness to the new equipment. To a degree, the problems noted here will play a part in the introduction of any “new” equipment:
► The new automatic-response EW equipment is a drastic change from the manual method of operation and evaluation. Consequently, the introduction of this new equipment meant that the senior EW specialists in the fleet would have to learn a whole new way of doing their jobs right along with the lower-rated men. As a result, some of them perceived the new equipment as a threat and remained highly skeptical throughout the tests. Since senior EW specialists frequently were on major staffs, their skepticism spread among fleet and type command staffs.
► Similarly, because the new equipment would mean a requirement for a new training syllabus in various fleet EW schools, and because instructors were unfamiliar with the technology and with how best to employ the new equipment, some of them felt threatened (at least by additional work) and tended to have a built-in bias against the new equipment. A more common reaction in the training community was simply to ignore the equipment in the hope that it would go away.
► To many fleet operators, deploying, participating in exercises, and simply being subjected to the trauma of everyday “flails” are the essence of the profession. The test ship was not generally subjected to these sorts of operations and, as a result, some fleet operators had a tendency to see whatever the test ship was doing as something smacking of the prima donna—and consequently not welcome in the fleet.
The OpEval was completed in the first week of September 1976. ComOpTEvFor’s final report predictably did not pick a winner but focused on pointing out the weaknesses of the systems as noted in the OpEval. In February 1977, the CNO Executive Board chose the Raytheon design, awarded a $1.6 million contract for follow-on testing and recommended to DSARC III that full-scale production be undertaken.[8]
Evaluation: Based on the fact that the Navy was in a position to make a decision regarding the purchase of a new series of EW equipments with a reasonably firm price, schedule controls, and a demonstrated level of performance, we can be reasonably optimistic about the positive effects of the new design-to-cost controls. Given the success of this program in controlling costs, it can be tentatively concluded that design-to-price controls will work at least up through the production phase if the following conditions prevail:
► The threat can be “frozen” so that later attempts to gold-plate can be avoided
► Off-the-shelf technology can be used
► The competitive nature of the testing can be maintained
However, it should be pointed out that this design-to-price EW program is far from being completed, and in any case one program does not prove the point. Some warnings are therefore in order:
► Controlling costs is different from controlling prices. Eventually the state of the art has to move ahead, and that will inevitably cost money. By freezing the threat and using only off-the-shelf technology, we may only be postponing costs rather than controlling them.
► A competitive source selection process is difficult to retain. An overriding concern throughout the tests was that one contractor might drop out, leaving the remaining contractor, instead of the Navy, in the driver’s seat. Establishing a source selection process wherein the two competing designs can be adequately tested against a standard was no mean feat under the best of conditions because of the inherent dissimilarities in the designs. Hence, despite everyone’s best efforts to maintain fairness, one can never be assured that the final selection of a contractor will always be totally objective.[9] The makeup of the source selection board, the personal preferences of key individuals at higher levels, the overall reputation of competing companies, and even follow-on imperatives may affect an otherwise objective choice based solely on competing designs or test results.[10]
► There will be pitfalls where control may be lost even after the production run has started. Despite the fact that the design-to-cost EW program progressed rapidly from commencement to source selection (49 months in all), the development portion alone outlasted one administration and two Congresses. Because of normal rotation patterns, Navy personnel will start moving out soon after the development phase is completed. In the technical community many of the key people tend to see a project as losing much of its challenge once it reaches full-scale production and thus will seek to move on to other new Projects, leaving the “details” to newer and less familiar individuals. Before long, only the contractors will be left as the most knowledgeable people still in the program. The most carefully laid plans and controls can quickly go astray when only the ones supposedly being controlled know exactly what ls going on. When a program moves from development to procurement, responsibility for its funding shifts from research and development to a combination of accounts. As a result, the corresponding roles of the key Navy participants change accordingly, thus providing a second point at which control may yet be lost.
A third such point is when the program shifts from production to installation. Planning for the installation is done by individuals who have had little or nothing to do with the program’s development and who will not have to use the equipment themselves. Unless they get some strong and specific directions from the operators, these individuals will invariably install the new equipments in the least expensive location, without considering crucial factors such as combat integration or follow-on equipment developments. This point is of particular importance in the EW system, for its function is qualitatively different from its predecessor, and a substantial portion of its gains in terms of processing time will be lost if it is not colocated with the combat-situation decision-makers. The ship’s tactical action officer (TAO) needs immediate access to all threat information as well as to all potential countermeasures, whether they are hard-kill such as guns and missiles or softkill such as decoys and maneuvers. The major advantage of the new EW suites is that the TAO can now be given timely threat information. But if the information has to be passed to him orally from some distant point, the time gained in threat warning and recognition will be lost in transmission, and inevitable confusion will result exactly when every second is most precious. The fact is that the older equipment which the new EW equipment replaces is not now generally colocated with the TAO, and installing the new equipment in the same (wrong) place will in the end only marginally improve the situation. In fact, installation planners are proceeding exactly as if the new equipment were simply an updated WLR-1, and no one has yet been able to right the giant step backward. Another problem with integrating the new SLQ-32 into the ship’s internal command and control network is the fact that it is not automatically tied in to combat systems such as radars, Navy tactical data system (NTDS), and command, control, and communication systems. Some developmental work has been done to overcome this deficiency, but no solution has yet been found which even approaches getting the full potential from the equipment.
Yet a fourth point at which control may be lost is in the introduction of the equipment into the fleet. New doctrines, tactics, and techniques will be required. Normally, these fall into the bailiwick of the fleet commanders, although most new equipment usually comes with recommendations from ComOpTEvFor as to some proposed techniques to be used. But existing biases on the part of school instructors and shipboard operators impede fleet-wide acceptance of the new equipment and the development of new tactics for its use. As a result, old doctrine and tactics designed for a different era with different equipment will probably continue to be used for a long time with this new equipment, and it may be years before the potential of the new equipment is fully realized. If these pitfalls can be avoided, this fleet operator, for one, will be extremely grateful and relieved.
Conclusion: The major test for the design-to-cost method will come on some multi-billion-dollar program in which the state of the art must be advanced. We will then know whether costs are really being controlled, or whether they are merely being postponed. If design-to-cost really does work, including its corollaries of manufacture-to-cost, procure-to- cost, and install-to-cost, then only one major problem will remain: research-to-cost. If that problem can be solved, the Department of Defense may at last have gained control over its own weapon systems acquisition programs. But in the end it will still remain for the military services themselves to obtain maximum utility from the products of that system, whatever the costs.
Lieutenant Commander McGruther was graduated from Dartmouth College in 1965 and received his master’s degree in political science (strategic studies) from Brown University in June 1978. He has served in the USS Loyalty (MSO-457), USS Hissent (DER-400), USS Roark (FF-1053), and was operations officer in the USS Leahy (CG-16) during the period when the ship was conducting the design-to-cost EW equipment tests. He has also had a tour with the Office of the CNO and was graduated with distinction from the Naval War College in 1974. He recently served on the Sea Plan 2000 study before reporting to his current assignment as executive officer of the Joseph Strauss (DDG-16). He is slated to report late this year to the Plans, Policy and Operations Division of the OpNav staff. Lieutenant Commander McGruther was the first winner of the Admiral Colbert Memorial Professional Prize Essay and has had articles published in the Naval War College Review and the Proceedings. Included was “The Dilemma of the U. S. Pacific Fleet (June 1978 Proceedings) which won first honorable mention in the General Prize Essay Contest. He is the author of a monograph, The Evolving Soviet Navy, published by the Naval War College in 1978.
[1] For footnotes, please turn to page 60.
1 Address by Deputy Secretary of Defense David Packard to the Annual Meeting of the Armed Forces Management Association, 1970.
[2] Address by Dr. John S. Foster, Jr., Director of Defense Research and Development, before the National Security Industrial Association, 12 March 1970.
[3] In “cost-plus” contracts the contractor promises to try to meet the performance requirements set forth in the contract within negotiated schedules and estimated costs in return for which he would be entitled to reimbursement of his costs, plus a profit.
[4] A “buy-in” is a situation in which a company deliberately offers a low bid to get its foot in the door and has the expectation of either raising prices later or making up in quantity for the low profit margin on individual items. In a “bail-out,” the government deliberately selects a contractor who has the greatest “need” rather than one offering the best product or lowest price.
[5] The council is comprised of the Director of Defense Research and Engineering; the Assistant Secretaries of Defense (Comptroller, Installations and Logistics, Intelligence, and Program Analysis and Evaluation); and the Director of Telecommunications and Command-and-Control Systems, as well as the applicable service secretary, JCS representative, and other parties specifically interested in the program being presented.
[6] CDR R. A. Hullander, USN, and John P. O’Brien, “Design-to-Cost: Reducing the Words to Practice,” Naval Electronics Systems Command Brochure, 1974, pp. 1-14
[7] Navy readers should consult NWP 33 for the list of which suites will be placed aboard which ships.
[8] Electronic News, 28 February 1977, p. 2.
[9] As recently as late 1976, in the case of the source selection of the Army s new XM-1 tank, also conducted under careful design-to-cost controls, the Secretary of Defense reversed the recommendation of the Source Selection Board primarily because he was more interested in commonality aspects than in proven test performance.
[10] The general concept of “follow-on imperatives” was explored by James R. Kurth in "Aerospace Production Lines and American Defense Spending in American Defense Policy, 3rd ed. (Baltimore: The Johns Hopkins University Press, 1973). Kurth’s thesis was that, as a major defense contractor nears the end of a large contract's production run, it is predictable that that company will be awarded another somewhat similar contract in order to “keep the line open.”