Document created: 18 August 03
Air University Review, March-April 1975

A Dynamic Process

DURING the past fifteen years many innovative techniques have been instituted to improve the development, procurement, and Support of defense systems. These improvements have been initiated in response to Outside pressure—Congressional and taxpayer concern with how We acquire systems—and as the product of dedicated individuals within the acquisition process whose efforts focus on the purchase of more defense per dollar.

In the 1950s and most of the 1960s we made decisions concerning the future of billion dollar programs based on analytical (paper work) analyses. When we progressed to production, we frequently encountered monumental problems, resulting in Cost overruns and slipped schedules. The fly-before-you-buy approach is a reaction to the problems associated with the total reliance on such analyses. In effect, we now test the hardware under consideration for purchase before committing ourselves to a production contract. (Obviously, this approach is not applicable when one or a few copies of an item are being bought.) Although development costs increase with inclusion of the hardware to be tested, we believe this additional cost will be more than offset by savings during production. Further, subsequent support costs will be reduced, since only one or a limited number of configurations of a system will be produced.

PIECOST

Overhead typically represents two-thirds of the in-plant expenses incurred by defense contractors. PIECOST, an acronym for "probability of incurring estimated cost," is a statistical technique designed to enable us to determine a measure of acceptable overhead cost.

With PIECOST we identify the variables that influence or drive overhead costs at each contractor's plant. Thus we are able to negotiate more realistic overhead rates on our large development and production contracts. Of equal importance, we are able to monitor the contractor's incurrence of overhead costs. If we note that the contractor’s costs are running high, we can inform him that he is spending at an excessive rate and direct that he correct the situation.

With the exception of World Wars I and II and the Korean conflict, we have attempted to purchase most of our supplies and services to meet defense requirements through formal advertising (competitive bidding). Prior to the 1960s many supplies and services could not be procured by use of formal advertising procedures, since we did not have specifications describing our requirements adequate to use with this method of purchase. In the early sixties, we developed a procedure called Two Step Formal Advertising to allow the benefits of formal advertising even when we did not have adequate specifications.

Under this procedure, we provide a brief description of our requirement and invite interested potential suppliers to submit technical proposals indicating how they would satisfy our requirement if they received a contract. The technical proposal does not give any indication of the cost involved for the approach being submitted. The technical proposals are reviewed and accepted or rejected based on predetermined criteria. After two or more technical proposals have been found acceptable, we proceed to step two in the process. In this step we invite the potential suppliers whose technical proposals have been accepted to submit formal bids on their respective technical proposals. In this manner we are able to get the benefits of formal advertising.

Since the time of the Revolutionary War, we have related profit under negotiated procurements to the cost of the goods or services being purchased. Until very recently, we tended to ignore the role of invested capital as a factor in developing profit objectives. We have recently developed a procedure designed to encourage defense contractors to invest in more efficient equipment and facilities.

Historically, there has been a high correlation between a contractor’s costs under a negotiated contract and the profits he achieved. This relationship places the incentive (as seen from the factor's point of view) on the side of inefficiency, for the greater the cost, the greater (in absolute terms) will be the profit.

Diametrically opposed to this approach is the European concept of basing profit solely on the amount of capital invested is approach guarantees a profit on a contractor's investment regardless of the efficiency with which he uses those resources. This is not our objective. We desire to reduce the inequities in the profit opportunity available and motivate the contractor to employ his resources efficiently. Under certain large dollar-value contracts we will determine our profit objective by giving equal emphasis to the contractor's use of his capital resources and his likely costs.

By increasing the emphasis placed on amount of contractor capital employed on a contract, we hope to reverse the historic incentive for contractors to cost. We are not going to the extreme of guaranteeing a profit based solely on the amount of capital invested. We are balancing between these two extremes. In the process, we hope to induce our suppliers to increase their efficiency and reduce their costs and ours.

Should Cost is a procedure used to determine what a system ought to cost, assuming reasonable economy and efficiency in the contractor’s operations. It represents a coordinated analysis of a contractor’s management, cost estimating, and production engineering procedures. The ultimate objective of the Should Cost approach is to provide the government with a more supportable negotiating position. This goal is accomplished by conducting an in-depth in-plant analysis and by challenging inefficiencies in the contractor's operation. The actual methodology consists of a five-phase program: planning, data acquisition, analysis, report, and negotiation.

The planning phase begins with the identification of a candidate for a review. The general criteria for selection are found in the following questions: (1) Is the program a major, ongoing one of high dollar value--$25 million or more? (2) Does the contractor have substantial amounts of negotiated government sales? (3) Has the contractor been operating in a sole-source atmosphere or another environment that does not require effective cost control? (4) Has there been substantial cost growth associated with the item being procured? (5) Will there be a significant number of follow-on production Contracts? (6) Does the planned award date allow adequate time for the review? And finally, does the project manager have a reasonable expectation of a payoff from the type of effort that goes into a major Should Cost analysis? Selecting the team members is the next step in the planning phase. The size of the team will vary with the magnitude of the effort. Generally, the team will have ten to thirty people, half of whom will be engineers. Team members must be highly capable, and great care must be taken during selection to insure that the proper balance of talent is obtained. The skills required generally include those of industrial engineers, designing engineers, production specialists, statisticians, accountants, cost analysts, management analysts, and any additional specialists required to analyze the company's product line (e.g., nuclear engineers, aerospace engineers, computer specialists). The planning phase concludes after the work has been apportioned to the team members and a master schedule has been established.

Phase two, the data acquisition phase, takes from one to four months. This is the actual on-site investigation of the contractor's operation. Before the investigation begins, however, the contractor must be briefed on the goals of the analysis team, to insure cooperation in gaining access to required information. Then every aspect of the contractor's operation is reviewed by the appropriate team members. The areas evaluated include plant layout, machine capacity and utilization, production scheduling and control, labor standards, make or buy policy, industrial engineering standards, quality control, general and administrative expenses, cost estimating, tooling, labor, production engineering, design engineering, engineering overhead, manufacturing overhead, and any other areas vital to efficient operations. These evaluations must be completely coordinated to assure that all pertinent facts are gathered without duplication of effort.

Analysis, the third phase, overlaps both the preceding and following phases. During this period the team members discuss and integrate their findings.

The report phase is the realization of the team's efforts. The report will be the basis for the government's position during negotiations. The report format is designed to make the report an efficient negotiating tool. The report contains suggested primary and alternative negotiation positions, findings, and recommendations.

The negotiation phase is the finale of the effort. The government is concerned with areas such as more efficient plant layout, better inspection and sampling techniques, and improved material purchasing practices, as well as the actual costs proposed for these elements. Individual team members contribute to these negotiations by providing expertise in the area they have evaluated.

The benefits of Should Cost are twofold: the short-term benefit of better pricing on the current requirement and the long-term benefit of more efficient contractor performance on future requirements.

While Should Cost is not a panacea, it has proven its effectiveness in analyzing high-dollar, major programs. As the method is further refined, its effectiveness should increase. The use of Should Cost principles to strengthen traditional analysis, coupled with the Should Cost review of selected major programs, will make detailed analysis more effective.

During the 1950s we found ourselves playing catch-up with the Soviet Union. Our concern was to develop high-performance systems—and quickly. Our management information systems provided cost information-after the fact. This information usually put us on notice that we were in a cost overrun situation after it was too late to take any corrective action. The net result of our 1950s approach to acquisition was high performance, slight program slippages, and huge cost overruns. Studies of acquisition during this period indicate the costs typically were 300 percent of those budgeted.

During the early 1960s we imposed new management information system requirements on our contractors. In most instances, the contractors found it expedient to resort to a sort of double bookkeeping—one information system for their information and control and another system to satisfy government requirements. Cost control in the early sixties was somewhat better than during 1950s, but still not good.

Now we have developed a concept that allows the contractor to satisfy both his and our cost and schedule information with one system. This approach is known as Cost/Schedule Control System Criteria (C/SCSC).

A prospective contractor must describe in detail how his management system works and the steps required, if any, to bring the system into compliance with C/SCSC. This approach requires that the system to be developed and produced be broken into a pyramid of units down to end components. It then relates the elements of work required to each other and to the end product. The contractor establishes completion schedules and target costs for each subunit and lower tier item C/SCSC recognizes that the contractor will have to reschedule and rearrange certain activities as the contract progresses and gives the contractor the freedom to make adjustments under certain constraints.

C/SCSC requires the contractor to have only one set of books for both his internal planning and control and for the required government reports. The contractor has to develop a program for work performance that the government can use to monitor the contractor's performance. In this context estimated and actual cost, schedule, and technical performances must be reported in summary terms to Department of Defense managers.

C/SCSC requires the contractor to assemble, review, and analyze the totals of several different categories of costs monthly. The results are then forwarded to the government for evaluation. A comparison is then drawn between the budgeted costs and actual costs. When actual costs exceed or fall short of the budgeted cost, a cost variance is detected which points out a cost overrun or underrun for an individual work package or for the contract as a whole.

Another comparison is drawn according to the time span originally established for the work packages in the planning and budgeting stage. If the items scheduled to be completed at a certain time have not been accomplished, an unfavorable schedule variance is noted. Quite possibly other items that were not scheduled to be completed are actually finished. This balancing factor may eventually erase the effects of schedule variance in future accounting periods. If any variance, whether in cost or schedule, is noted through the C/SCSC analysis process, the contractor is required to trace the cause of the variance. He must determine and explain its origin and the steps he is taking to correct any deficiency.

Our experience in the 1950s and early 1960s demonstrated the need for an integrated system of inspection and evaluation of engineering requirements, cost, and schedule performance that could provide greater visibility for the program manager. C/SCSC has been designed to provide a means of comparing actual schedules and costs with budgeted schedules and costs. Thus we now have the capability to analyze problem areas in time to take effective management action. C/SCSC serves to flag problem areas that do not fall within the parameters of acceptable variance. Through the principle of management by exception, it focuses only on those items that cause a variance in cost or schedule which may result in a revision to desired product performance. Because of the timely reporting of information, danger areas are now detected early enough to afford a direct resolution before the errors compound. The contractor is required to take the initiative to solve these problems and, at the same time, report the problems and their selected solutions to the government for further analysis.

The complexity of today's defense systems and the desire to be ahead of other nations in technological achievement frequently require a contractor to research deeply and develop processes that were beyond the state of the art when the contract was awarded. The nature of this work lies in uncertainty, which breeds variances from time schedules and planned costs. Thus C/SCSC is not expected to eliminate cost growth. But management now has an effective tool to project the results of schedule and cost variance on the desired system and provide a sound basis for the decisions necessary to limit cost growth.

The award fee concept is a relatively new innovation. It is an extremely flexible approach to contracting that improves communications within the buying and the selling organizations and between buyer and seller. The award fee provision is the major element, in terms of contractor motivation, in the Cost Plus Award Fee (CPAF) contract and a significant factor when the award fee is used in conjunction with other types of contracts. It is specifically designed to provide an incentive to the supplier for superior Contract performance. The philosophy behind the award fee is to give the supplier a monetary incentive and to give the government a flexible management tool with which to influence performance. The award fee itself is simply a "fee pool" (a specific dollar amount) established by the buyer (DOD) and awarded to the supplier (contractor) on a periodic basis. The amount of this fee pool that the supplier can earn is dependent on his performance—as determined unilaterally and qualitatively by us, the buyer—over and above the minimum requirements set down by the contract. It is possible for the contractor to earn from zero to 100 percent of the award fee, dependent, of course, on his performance.

The award fee has a periodic aspect in that its entire dollar amount available for the life of the contract may be broken down for disbursement at specific evaluation periods. Formal performance evaluations are made periodically over the life of the contract, and a fractional award fee determination and payment is made at the end of each evaluation period. For example, if a contract were awarded for a one-year period with a total award fee of $1,000,000 and quarterly formal evaluation and fee disbursements, the supplier might typically be offered $250,000 during each of the evaluation periods as the maximum amount he could earn. Formal evaluation and payment would be made at the end of each quarterly period.

The periodic nature of the award fee concept allows us to make a thorough evaluation of progress, make necessary changes in areas where conditions have changed or performance is not as expected, and provide useful feedback to the contractor on how he is progressing. The flexibility in the award fee provision stems essentially from four factors: (1) the subjective nature of the performance evaluation and fee determination process, (2) our right to change or modify areas to be considered for performance evaluation from one period to another, (3) the versatility with which the amount of the award fee can be distributed over the life of the contract, and (4) the fact that no absolute requirement exists to have specific evaluation and fee determination periods set for all contracts. This inherent flexibility represents a major advantage of CPAF contracting over other types of arrangements (e.g., firm-fixed price, fixed-price incentive, cost-plus-incentive fee, cost-plus-fixed fee, and various multiple incentive contracts).

Increasing use of the award fee concept is being made in DOD acquisitions. Much research and development work is procured by use of this technique. The operation and maintenance of the Air Force Arnold Engineering Development Center are in the process of being accomplished under an award fee contract. The award fee concept has been used with excellent results in conjunction with other pricing techniques in the acquisition of the F -15 aircraft.

One of the more recent developments in the acquisition field is called "design-to-cost." Although this concept has been used by industry for years, its application to the acquisition of defense systems is relatively new.

The incremental acquisition strategy that DOD has adopted can be compared to the product development process used in private industry. Generally, when a private firm initiates a new product, it first assigns that product to a small team of personnel from engineering, manufacturing, and marketing. The team develops estimates of required technology, market impact, and manufacturing requirements as well as their associated costs. Executive management reviews the project team's findings at specified intervals during the development process. At each of these review points, anticipated price and return on investment are compared with the expected cost of production. The program proceeds to the next stage of development if projected production costs and net revenue are satisfactory. If they are unsatisfactory, alternatives are examined for cost correction or the project is terminated.

With the presently established procurement policies, DOD follows a similar process for system development and acquisition. A proposed major system is subjected to several stringent reviews by the military department, the Defense Systems Acquisition Review Council (DSARC), the Office of Management and Budget (OMB), and Congress before each funding milestone. The decision for continued development requires satisfactory findings as to expected system performance and projected system costs. Consequently, the total costs of a program must be commensurate with performance and must fall within budgetary constraints. This may require trade-offs in system performance and schedule. . The strategy requires that viable alternatives be maintained until such time as the system selected for development has demonstrated the required performance and supportability within cost constraints. 

This concept of procuring systems within a cost constraint has been given various titles, including design-to-cost, design-to-price, cost-to-produce, and design-to-cost-to-produce. For simplicity, we refer to the concept as design-to-cost.

Use of the concept requires the establishment of a unit production cost we can afford to pay for the quantities needed. The unit production cost is a primary design parameter equal in importance to system performance parameters. The concept requires that cost be emphasized continuously in trade-off decisions and that the contractor demonstrate his ability to achieve the cost target before award of the production contract. Use of the concept requires attention to four key elements: (1) system cost targets, (2) system performance goals, (3) production plans, and (4) feedback mechanisms.

Establishment of the cost target is probably the most crucial aspect of design-to-cost. In the first or conceptual phase of a system acquisition, the total estimated future cost of a program depends on the technology required, the number of units required, monetary inflation, delays, changes in system performance characteristics, and numerous other cost factors. Since some of these factors can be estimated only imprecisely, cost estimation techniques are extremely important in the establishment of the initial target cost.

The establishment of system and subsystem performance goals is another important factor in the use of the design-to-cost concept. Unlike some acquisition policies, however, performance is not the dominating characteristic for program evaluation. Though desired performance characteristics may be specified, the concept requires an ability to trade performance factors for greater cost savings consistent with some minimum levels of performance. It should be noted, however, that failure to meet the cost target or the minimum performance levels would require that the program be examined for possible alternatives or termination. Consequently, design-to-cost requires rigorous use of cost-benefit analysis. Increments in cost must be justified by the benefits derived in performance from proposed system or subsystem designs, materials, or production methods. These cost increments must be consistent with the overall cost target.

System design not only influences performance characteristics, reliability, and maintainability; it also influences the type of production method to be used. In addition to system design, the number of units required plays a major role in determining the production process and consequently, the unit production cost: Design and the quantity required Specifically affect direct labor, direct material, and factory overhead. For example, designs requiring special tooling or "clean room" facilities will increase a manufacturer's overhead. Usually specialized equipment is more economical for large production runs, and general-purpose tooling is ordinarily more cost-effective for smaller runs. The variable and fixed costs associated with labor, material, and overhead will vary depending upon the processes specified during system design. 

When the Secretary of Defense approves the request to enter full-scale development, the unit production cost is established and becomes a firm requirement of the development contract. During development, it is essential to track the designs of those items comprising the significant cost elements of the system. Though these items are only a small part of the total system, they comprise the major portion of total system cost. Historically, 20 percent of a system's components constitute approximately 80 percent of its cost.

By monitoring design progress of the major cost components and evaluating the effect of designs on production costs, one can determine the need for redesign action to meet the design-to-cost goal. Figure 1 depicts an example of a system's projected production costs based on an analysis of designs produced. The top portion of the figure illustrates progress toward the unit production cost goal through successive design iterations and shows how the impact of early system designs can be extrapolated to reveal potential production cost overruns. The lower portion shows the expenditure of the development budget as design iterations and time progress. If the possibility of an overrun exists, alternate designs should be developed. These design iterations of the major cost items should occur early in the development phase of the system program. Otherwise there may be insufficient development funds remaining to correct designs, which could cause excessive production costs. Thus, early design review will usually prevent sunk costs from consuming a major portion of the development budget. Further, the early review of designs is important since system design will ultimately influence not only cost but also performance, reliability, and maintainability.

As has already been noted, design-to-cost is part of the overall DOD incremental acquisition strategy. In an effort to provide system programs with exercisable, viable alternatives, the strategy calls for the Defense Systems Acquisition Review Council (DSARC) and service reviews throughout the life of the development program. Coupled with these reviews is the requirement for separate contracts for development and production. During the DSARC and service reviews, cost is given major consideration for program continuation and is the basis of the ultimate decision to enter a production contract. Though system performance goals are specified, design-to-cost requires trade-offs in performance and schedule to meet the cost objective consistent with these stated performance requirements.

Our experience with design-to-cost is very limited. There is a belief on the part of several practitioners that the design-to-cost approach is most applicable at the subsystem and even component level. As an observer of the evolving acquisition process, I have little doubt that proper application of design-to-cost will result in significant savings of DOD acquisition and operating dollars.

Perhaps the most significant improvement in the acquisition process is the introduction of the Integrated Logistics Support (ILS) concept. During much of the 1950s and 1960s our approach to system acquisition was predicated on the principle of concurrency—concurrent system development, test, production, and even operation. The objective was to achieve an operational capability at the earliest possible time. Engineering and design changes were made on equipment in production and even after deployment, to correct deficiencies revealed during testing of the system. The concurrency concept created severe problems for those personnel subsequently supporting and maintaining the system. Spare parts, test equipment, and technical data ordered during previous months were not always compatible with the end items being produced during the current month. Under this concept performance and early operational capability were emphasized, frequently at the expense of future support costs. The reliability and maintainability of subsystems and components were not given the same attention as was system performance. Yet, as indicators of how long an item would perform satisfactorily under stated conditions and how quickly an item could be repaired when it failed, reliability and maintainability have tremendous implications when considering how much a system would cost to own. Frequently we found that an extra dollar invested during development or production would save us ten dollars during operation of the system. Unfortunately, under our concurrency concept of acquisition, too little attention was paid to the support implications of the chosen system design.

During the past few years the ILS concept has been developed and implemented. Under this approach, we now provide visibility of the support requirements essential to perform system tradeoff analyses. Thus we can reduce the total cost of ownership of a system, including the cost of both acquisition and support during operation. Under its, we have assigned highly qualified logisticians to the project offices responsible for acquiring new defense systems. These logisticians cause us to analyze support implications of each approach under consideration. ILS should significantly reduce the total cost of owning a system.

IN HIS recent book Arming America,* J. Ronald Fox found much wrong with the acquisition process. I concur that there is a need to make many changes, but I also believe, as reflected in this article, that those in the defense acquisition business have not been idle. Much has been done, and we are actively working on the much more remaining to be done.

*This book was reviewed by Colonel Burt in January-February 1975 issue of Air University Review.