The long, drawn- out and expensive acquisition process in the military is a serious problem at any time, but especially so in these times of tight and declining budgets. Many times the Department of Defense (DOD) has been criticized for the manner by which it procures new weapons systems. Recent articles in the Washington Post1,2 and Defense News 3 review the Army acquisition process, presenting an analysis of some $38 billion spent on system s that were terminated for high costs and lengthy development times. Two cases in point that f ailed: the Crusader artillery system which foundered, in part, on the failure of a new propellant system , high costs, and a too-long development cycle, and the Comanche stealth helicopter, again terminated because of cost overrun s and slippage of schedules. The available funding for the Comanche was redirected into buying and upgrading existing helicopter program s where the procurement risks were reduced. A question is how much of this problem is due to technical inefficiencies, and how much is due to other factors such as the requirements process, Defense acquisitions regulations, and the complexity of the lengthy budgeting process or changes in the governing political system . We seek to show how one factor, the expanded use of high performance computing, can contribute to improving the design and production of weapons systems. This would contribute to a more responsive, more economical acquisition process.