Quantifying the effects of commercial processes on availability of small manned-spacecraft

Abstract

The methodology described herein identifies and quantifies equipment and system level failure modes based on the criticality of their effects on system functionality. This methodology is useful for systems that require high-reliability assessments, such as, spacecraft that are developed with "faster, better, cheaper" commercial processes. The authors developed this methodology by integrating the similarity/failure cause analyses methods developed by the International Electrical Commission (1999) with the process grading methods developed by the Reliability Analysis Center (1998). Since the advent of Acquisition Reform in 1994, the authors have studied the effectiveness of many of the "streamlined" reliability assessment techniques used in military space programs. What they learned is that every method can be of some use in identifying, mitigating or estimating reliability risk, but selecting the minimal set of methods for a high-reliability assessment requires looking beyond task names. Management organizations must understand how the product-performance objectives are supported by the methods used. Based on the authors' experiences, most spacecraft manufacturers do not practice performance-based reliability assessment methods, and yet they successfully meet or exceeded the predicted availability/reliability of their systems. However, the few satellite and launch vehicles that failed in 1998 and 1999 resulted in billions of dollars of financial losses and managerial shakeups at some major corporations. In general, public opinion is tolerant of these kinds of losses because they are perceived as the cost of doing commercial business in space. That is not the case when failure of a manned Space Shuttle transport occurs. Over the next few years, the commercial spacecraft industry will develop small to medium-size, single-stage manned-spacecraft. Organizations will have to use methods for identifying, mitigating and predicting critical failure modes more accurately than those currently used for unmanned-systems.