System Testing Reveals Problems in the Kill Vehicle Program

On June 23, 2014, the U.S. Missile Defense Agency reported long-awaited good news. Its Exoatmospheric Kill Vehicle (EKV) Capability Enhanced II kill vehicle successfully intercepted a missile fired from the Marshall Islands. The kill vehicle, a warhead launched from the Vandenberg Air Force Base in California, hit the intercontinental ballistic missile in midcourse, that is, after it had been launched and prior to its reaching the target. Although the engineers at the air force base left little to chance by programming the exact coordinates of its target into the kill vehicle, the test was considered an important success. The three previous attempts, all executed between 2008 and 2014, had failed. In fact, of the 17 tests of the Ground-based Midcourse Defense (GMD) system, only eight had hit their targets, giving the project-which will cost American taxpayers $40 billion by 2017-a 47 percent success rate.

In 1983, President Ronald Reagan first backed the idea of creating missile defense technology that would render nuclear weapons obsolete. The Strategic Defense Initiative (SDI), popularly known as Star Wars, cost taxpayers approximately $30 billion before the project was abandoned. In 1999, however, the U.S. Congress decided that it would begin development of the GMD system to protect the United States against a nuclear missile launched by a rogue nation. In 2006, communist North Korea had launched its first successful test of a nuclear missile, and both Iran and Iraq had made efforts to obtain nuclear technology. Congress set no date for the completion of the GMD system, but simply specified that it should be deployed as soon as it was technologically feasible.

However, terrorists attacked the Twin Towers and the Pentagon on September 11, 2001. On December 16, 2002, President George W. Bush issued a directive to deploy the GMD system by 2004, although two tests of the system had failed a week earlier. Rather than continuing work on the prototype and test the system, the Missile Defense Agency began deploying missiles in silos. Approximately 38 percent of the kill vehicles' software had not been validated through flight testing. The political pressure to deploy in the wake of the 9/11 attacks overrode concerns about faults in the software and other components.

In 2004 and 2005, however, the Missile Defense Agency aborted tests of the GMD system when the missiles remained stuck in their silos due to software failures. In 2010, a test failed because the sea-based radar system the kill vehicle used to calculate the trajectory of the transcontinental ballistic missile became confused by pieces of metal that tore away from the missile as it approached the kill vehicle. Before each new test, engineers redesigned the defective component. However, the Missile Defense Agency has stopped short of a total redesign and carefully planned systems development of a new prototype.

In February 2013, Frank Kendall, the undersecretary of Defense for acquisition, technology, and logistics, became the first to admit to the inherently failed development process when he announced, "The root cause was a desire to field these things very quickly and very cheaply.... We are seeing a lot of bad engineering, frankly, and it was because there was a rush."

Clearly, if the kill vehicles can only hit the missiles when they are given the coordinates in advance, these warheads will likely fail to bring down a nuclear missile fired by a rogue nation when both the launch time and position of the missile is unknown. The question remains as to whether the United States will be able to develop a functional GMD system before a rogue nation or terrorist organization becomes capable of launching a nuclear missile that can reach the United States.

Discussion Questions
1. What error did the Missile Defense Agency make in the development of the GMD system?

2. How do the factors that pushed the agency to release the GMD compare to those that rush the development of business software?

Critical Thinking Questions
1. What steps can businesses take to ensure that engineers report all potential problems that arise during development?

2. What steps can IT professionals take to make business managers aware of the importance of following the complete systems development process?