The intent of CAIV is to provide the customer/warfighter with highly capable systems that are affordable over the life cycle. The CAIV process is twofold. First it is essentially a planning activity establishing and adjusting program cost objectives through the use of cost-performance analyses and trade-offs. This is facilitated by a Cost Performance Integrated Product Team (CPIPT) during all phases of the acquisition program. Organization and activities of the CPIPT are described in DoD 5000.2-R for ACAT I programs. The principles outlined in DoD 5000.2-R may apply to ACAT II & III programs as well.

The second component of the CAIV process involves execution of the program in a way to meet or reduce stated cost objectives. Requests for Proposals (RFPs) and contracts should communicate cost objectives and incentivize industry to meet or better them. These cost objectives, much like performance requirements, will likely be flowed down/allocated, by the contractor, to the lower levels of implementation/design. It will be up to the contractor, as part of the design process, to conduct necessary cost-performance-schedule-risk trades as appropriate to produce a system that meets overall contractual requirements. Contractual implementation of CAIV can be facilitated by a systems engineering cost control process in designing the system. It is important that a contractor's detailed design and cost target flowdown activities are consistent with approved program CAIV objectives. As with any process, appropriate metrics should be devised for tracking progress in achieving cost objectives.

Early in the acquisition cycle, the high leverage of CAIV inspired cost/performance/schedule trade-offs should be shaping the requirements and proposed design approaches on a cost-effectiveness basis. Once a specific design is chosen, overall cost objectives should be allocated to specific cost and system elements. It is at this time that cost objectives should be translated into design requirements and that DTC techniques come into play. As the program approaches production, funding constraints are better known and budget begins to dominate the tradeoff ground rules. Cost-effectiveness will be modified by affordability considerations as the trade-offs start to focus on the cost-effective alternatives that are practical from a budget point-of-view. Monitoring cost-sensitive characteristics of important system performance and manufacturing processes can give early warning of a production cost growth during development and production.

Military hardware development programs have historically experienced moderate cost and schedule growth. Some typical contributors to this have been: (1) a performance-dominated requirements specification process that begins early and is rigidly maintained through much of the development, (2) a development process that is performance-driven, (3) a design that is near the feasible limit of performance that can be achieved at a given time, (4) uncertain and optimistic assessment of the feasible limit of performance that can be achieved in a design for a given cost and schedule [1].

The common thread running through these problem areas is that the design is "pushed" in the direction of performance, which often translates into unexpectedly higher levels of program cost, schedule and risk. The proper application of CAIV principals early in the development process can potentially alleviate the underlying problems that lead to cost and/or schedule growth through: (1) identification of a range for key requirements (e.g., threshold and objective), (2) trading among requirements, (3) reducing the dominance of a performance driven design, and (4) recognizing the level of risk present and implementing a proactive risk management process.

I will now illustrate how several of these factors can lead to increased program cost and schedule through an unrealistic design. (This design could, hypothetically, result from performance-dominated requirements specification or a development process that favors performance.) Price versus performance data was collected for a common microprocessor from a vendor in April 1996 and is plotted in Figure 1. The circuit price varied solely with a single measure of performance--microprocessor clock rate (speed in megahertz (MHz)). In this case, the last 11 percent of performance (150 to 166 MHz clock rate) leads to a 45 percent increase in processor price ($469 to $678 dollars).

If the development phase design is near the upper limit of achievable performance at a given time (166 MHz here), then significant increases in development and production phase cost and/or schedule can result from a small increase in performance. Similarly, a design in the region of the upper limit of performance often entails considerable risk that must be proactively managed through a viable risk handling process. Note also that backing off slightly in performance can lead to a substantial decrease in cost and/or schedule. As previously mentioned, the last 11 percent in performance in this case leads to a 45 percent increase in price. In many cases the last 3 percent of performance that can be obtained at a given time can lead to a 20 to 30 percent increase in price. Consequently, it is important for engineers and managers to identify where the "knee of the curve" is for cost/performance/schedule/risk relationships and actively use this information in the design process. This concept has historically been used in commercial design practices as well as in the former Soviet Union's military programs far more effectively than in U. S. defense programs. However, if CAIV is properly implemented, at least some reduction in defense development program cost and schedule growth as well as risk should be achieved.

[1] Edmund H. Conrow, "Some Long-Term Issues and Impediments Affecting Military Systems Acquisition Reform," Acquisition Review Quarterly, Defense Acquisition University, Summer 1995.