FY97 DOT&E Annual Report


Air Force ACAT IC Program
One-of-a-kind system
Total program cost (TY$) $1746M
Average unit cost (TY$) $342.5M
Full-rate production 1QFY89
Rebaselined 4QFY94
Integrated Mission IOT&E 1QFY99

Prime Contractor


The Cheyenne Mountain Upgrade (CMU) Program consists of upgrades to North American Air Defense Command's (NORAD) ballistic missile, air, space, and command center elements within the Cheyenne Mountain Complex outside Colorado Springs, CO. CMU also upgrades and provides new capability to survivable communication and warning elements at the National Military Command Center (NMCC), U.S. Strategic Command (USSTRATCOM), and other forward user locations. CMU additionally provides at Offutt, AFB an austere backup to Cheyenne Mountain ballistic missile warning.

CMU's modernization and new capabilities enhance the ability of our nation's civilian and military leadership to rapidly and accurately correlate Integrated Tactical Warning and Attack Assessment (TW/AA) information in direct support of their critical need for information superiority. CMU provides commanders with timely, accurate, and unambiguous information regarding enemy missile, air, and space systems throughout the full spectrum of conflict. CMU allows for immediate response to changing situations and the military advantage of full-dimensional protection.

The CMU program consists of the following major subsystems:


During World War II, the national attack warning system was developed to warn the United States of an air attack. In the postwar era, this system and similar Canadian systems were brought together as the multinational NORAD. With the development of large computers and with the advent of the threat of ballistic attack, NORAD was provided with a series of semiautomated warning and assessment systems, culminating with the development of the 427M system. These systems were contained within Cheyenne Mountain. The 427M system became operational in 1979.

Beginning in 1979, NORAD and the U.S. Air Force Air Defense Command developed a series of command-level programs to resolve operational and sustainment problems with the 427M system. Problems were addressed by the creation of individual acquisition programs with limited scope and cost. By the mid 1980s, six programs were underway with an aggregate cost of almost $2 billion. It soon became evident that the set of CMU upgrades needed to be restructured as a single, major, integrated upgrade program. The consolidated CMU program was formally started in 1989.

CMU was placed on the DOT&E oversight list in 1990. In 1991 a test concept centered around end-to-end tests was approved for each of the three missions: air defense, space control, and missile warning. Several subsystems were tested between 1991 and 1993 with mixed results. Subsequently, three IOT&Es scheduled for 1993 had to be delayed or suspended because of software problems. Two IOT&Es conducted in 1994 identified serious performance and integration problems; and CMU breached its Acquisition Program Baseline. DOT&E assessed the CMU program in 1994 as neither operationally effective nor suitable.

The Air Force restructured the program in 1994 with a three-year stretch-out (from 1996 to 1999) and increased funding by $48M. The OT concept was revised to lengthen the test program and to use a series of combined DT/OTs to support incremental, time-phased acceptance decisions based on each subsystem's operational maturity.

An end-to-end IOT&E of the missile warning mission was conducted in 1996. This was one of the largest OTs ever conducted, linking a worldwide network of 10 radar and infrared satellite sensor sites, the Missile Warning and NORAD Command Centers within Cheyenne Mountain, the Alternate Missile Warning Center, and over 25 worldwide forward user sites, including the National Military Command Center. Based on this test, DOT&E assessed the CMU subsystems supporting the missile warning mission as operationally effective and suitable, with some limitations. The operational community determined that these limitations had no significant operational impact as workarounds, and near-term fixes were already in place.


The two major OTs conducted in 1997 were a test of Granite Sentry and an end-to-end test of the Air Defense mission. The CMU subsystems under test included CSSR, SCIS, Granite Sentry, CCPDS-R, and all of their interfaces.

The two tests were conducted simultaneously and included eight scenario test days during which 16 scenarios were run. Air mission scenarios were injected directly into the Communications System Segment Replacement front end. To test the nuclear detonation mission, simulated detonation messages were injected into the Defense Support Program Data Distribution Center. Crew operations were observed at the Air Defense Operations Center, the NORAD Command Center within Cheyenne Mountain, and at forward-user command centers.

IOT&E scenarios generated up to 120 aircraft tracks at a time, which Granite Sentry consolidated to approximately 70 tracks. These scenarios generated approximately 250,000 messages that had to be traced through the system, collected at various points, reduced, and analyzed.


The Granite Sentry and Air Mission OAs were not rated by AFOTEC for effectiveness and suitability. However, based on direct observation and on data provided, DOT&E determined these subsystems to be both effective and suitable.

All key operational requirements were met and Cheyenne Mountain Operations Center operators were pleased with Granite Sentry. While Granite Sentry's successfully achieved its primary functions of message processing, video distribution and computer display, it did not adequately perform its secondary role of supporting nuclear detonation operations. This deficiency has been corrected. A strong plus for Granite Sentry was that it had no downtime--a problem that had plagued the previous Granite Sentry and other CMU subsystems.


Some problems were found in processing nuclear detonation data that resulted in incorrect data being displayed by Granite Sentry. This deficiency, which was fixed and retested in March 1997, involved the Integrated Correlation and Display System. If not found and fixed, this deficiency may have affected decision making at the highest national levels. The test team found a tenfold overestimate of nuclear yield that may otherwise have gone undiscovered. An error of this magnitude could have resulted, at the very least, in ambiguity in the data provided to key decision makers. At worst, this error may have led to miscalculations in our national military posture with potentially serious consequences.

Air defense situation awareness was found to be a problem at some forward locations. Questionnaire results from the National Military Command Center and Canadian National Defense Operations Centre indicated that the Processing and Display Subsystem did not provide adequate visual displays to achieve the desired overall situational awareness. Based on these results, 18 enhancements were identified, several of which will be implemented in the winter 1997 software update. For example, the number of displayed air tracks will be increased from 10 to 50, some icon sizes will be reduced, and on-screen database management functions will be improved.

The integrated CMU OT&E is scheduled to begin in April 1998.