The Clinton administration will use several criteria to decide next July whether to go forward with deployment of the planned National Missile Defense (NMD) system. One of these criteria is the technical readiness of the system for deployment, which the Pentagon will assess in a Deployment Readiness Review (DRR) to be conducted in June 2000. The DRR will be based in large part on the results of three intercept tests that will take place before June 2000; the first of these is scheduled for Saturday, November 2. But, as we show below, the Pentagon will not have enough information to assess the technical readiness of the NMD system by next summer.

Criteria for Technical Readiness

For any system-military or commercial-to be technically ready for deployment, it must meet all three of the following criteria. It must be

1. technologically mature
2. operationally effective
3. reliable
   
   

A rigorous test program is the only way to assess whether a system meets these three criteria.

The Pentagon's Planned Test Program

The Pentagon is planning to conduct 19 intercept tests prior to completion of NMD deployment in 2005, but at most three of these will take place before President Clinton makes a deployment decision next summer.

Intercept test #1 will test the exo-atmospheric kill vehicle (EKV), not the full system. It will use a surrogate booster, and none of the other system components will be integrated into the test. Instead, to simulate the information that would normally be provided by the radars, a GPS transmitter on the target (with a C-band transmitter backup) will provide target location to the booster to allow it to dispense the EKV in the correct place. After it is dispensed, the EKV will operate on its own and is intended to home on the target using its infrared sensor. In addition to the target, the test will also include one balloon decoy, according to BMDO.

Intercept test #2, currently slated for January 2000, will incorporate the ground-based radars, but the first integrated system test (to include all the components) will be intercept test #3, scheduled for April 2000. However, intercept test #3 will still use a surrogate booster. The first test to use a prototype booster will be intercept test #5, scheduled for 2001.

What will the Pentagon's test program tell us?

What might -- and what can't -- the United States learn about the three aspects of technical readiness from the test program by next summer? Or by 2005, the target date for completion of deployment?

• Technological Maturity

  To assess the maturity of the basic technology, a prototype of the full system must be tested under controlled conditions. If all three intercept tests prior to June 2000 are successful (and we expect that at least some of them will be), it would be a strong indication that the basic technology is in hand. But these tests would still not provide an adequate assessment of technological maturity of the system since only one of these tests will be a full system test, and a surrogate booster will be used even for that test. Such an assessment should be possible by 2005, since additional full-system tests using prototype components will be conducted by then.

• Operational Effectiveness

  The operational effectiveness of a system refers to its expected performance under real-world conditions, as compared to controlled test conditions. In particular, the real world performance of the NMD system will depend critically on the countermeasures that missile states employ to defeat the NMD system. To assess the operational effectiveness of the system, it must be tested against a wide variety of countermeasures that approximate as closely as possible those that would be available to emerging missile states.

  What countermeasures will be available to emerging missile states? According to the 1999 National Intelligence Estimate (NIE), released in September, "Many countries, such as North Korea, Iran, and Iraq probably would rely initially on readily available technology-including separating RVs, spin-stabilized RVs, RV reorientation, radar absorbing material (RAM), booster fragmentation, low-power jammers, chaff, and simple (balloon) decoys-to develop penetration aids and countermeasures." Moreover, the NIE states, "Russia and China each have developed numerous countermeasures and probably are willing to sell the requisite technologies."

  For the NMD system, the difficult task is not detecting all the objects, but discriminating the real warhead from the false decoys, especially if anti-simulation techniques are used to make the warhead appear similar to the decoys, or all of the objects appear different from one another. If the NMD system does not know what the warhead will look like, it will be difficult to distinguish it from the decoys.

  Will the Pentagon test against realistic countermeasures before next summer? No. In the first test, a balloon decoy will be used. This decoy and those to be used in the next four intercept tests will help the Pentagon assess whether the EKV can perform the basic task of using its infrared sensor to distinguish a warm warhead from a cool object. No one doubts that this is possible. However, to assess operational effectiveness, quite different intercept tests would need to be done. Tests of realistic countermeasures would, for example, use a cooled warhead or decoys heated to different temperatures, or have all the objects enclosed in balloons of different surface coatings that would heat up to different temperatures in the sunshine.

  In fact, the claims of BMDO officials that the upcoming tests will use realistic countermeasures are contradicted by the Pentagon's Director of Operational Testing and Evaluation (DOT&E). In discussing the first intercept test with Inside Missile Defense (August 25, 1999, p. 13), Brig. General Willie Nance, the NMD JPO Program manager, "characterized the target suite as 'more than representative' of the decoys and countermeasures that a rogue state might employ." However, in its most recent (FY 1998) Annual Report, the DOT&E states that "The NMD ... program is building a target suite that .. may not be representative of threat penetration aids.... Test targets of the current program do not represent the complete "design-to" threat space..."

  Will the Pentagon test against realistic countermeasures by 2005, when initial deployment would be scheduled for completion? It is not clear. Beginning with the sixth intercept attempt, the countermeasures used will reportedly be more capable. However, whether these countermeasures accurately represent the real-world threat remains to be seen. Yet tests using anti-simulation and realistic countermeasures are critical to understanding the operational effectiveness of the planned NMD system, and should be conducted before a deployment decision is made.

• Reliability

  Because any incoming missiles would likely be armed with weapons of mass destruction, the Pentagon has reportedly set as its goal that the NMD system have a kill probability against an incoming missile of 95%, with a 95% confidence level. The Pentagon reportedly assumes that each interceptor will have a "single-shot" kill probability of 85%, so it plans to fire four interceptors at each warhead.

  Why is it necessary for the US military and political leadership to understand how effective the NMD system will be? First, the military must know the single shot kill probability to decide how many interceptors to fire at each target. Second, if the system is to have any utility against a deliberate attack by an emerging missile state, the US political leadership must know how big a risk they are running that a warhead targeted at a US city will get through.

  If the United States deploys an NMD system, it is reasonable to assume that it will be used very rarely. Thus, to assess the reliability and effectiveness of the system, the production version (not a prototype) must be tested enough times to give statistically meaningful data. This will not happen by next summer, or even by 2005.

  Military systems that require destructive testing are rarely tested enough before deployment to give a statistically significant measure of their reliability. Additional information will be gained for many of these systems through training and combat experience. However, the NMD system will require a much higher confidence level than typical military systems, and the United States cannot afford to wait for combat experience to assess its reliability. Yet, ironically, due to the compressed development schedule, the Pentagon expects to test the NMD system less than a typical military system.

The Pentagon's Criteria for Technical Readiness

The Pentagon has announced that it will give a green light to deployment once two intercept tests are successful, provided one is a full-system test (recall that intercept tests #1 and #2 will not be full-system tests). To lower the bar even further, it will give a provisional green light if only one of the intercept tests before June 2000 is successful, with deployment slated to begin in Alaska once another intercept occurs. Thus, if the first intercept test is successful-even though it is not a full-system test-next June the Pentagon will declare the NMD system to be technically ready for deployment, provided another intercept occurs later.

The Bottom Line: The Technical Readiness Decision Is Not About Technology

Next summer is too soon to make any meaningful deployment decision based on the NMD testing program. The Pentagon will simply not have enough information to assess the technical readiness of the system. In fact, the planned test program will not allow an assessment by 2005.

Originally posted September 24, 1999