Developing effective reliability growth strategies for DoD programs

Abstract

Planning for reliability growth requires a disciplined approach rather than an “ad hoc” approach. The reliability growth program must be devised early so that adequate test phase lengths and test assets can be planned to optimize reliability growth opportunities. The primary goal of the Reliability Growth Test (RGT) is to improve the probability of demonstrating achievement of the reliability requirement during Operational Testing (OT); which should also translate to effective reliability performance when fielded. The first step in the planning process is to determine the length of test time needed to reach a planned reliability growth goal before entry into OT testing. The planned growth goal should be above the reliability requirement to ensure that the requirement can be demonstrated with at least an 80 percent statistical lower confidence bound (LCB) during the OT phase. The primary inputs to planning models in general are: the initial reliability value from which the growth will commence, the anticipated reliability “growth rate”, the target or goal reliability, and the planned test length (phases). The initial starting reliability value is a critical input variable as it tends to influence overall test length. The best method for selecting an initial reliability value is to have some early test results where reliability has been determined by actual observed performance. The alternate method of selecting the initial reliability value based on some percentage of the requirement would entail formulating a less effective and higher risk test strategy. The next value that is critical to planning models is the growth rate. Choosing growth parameters that are realistic is an important step so that adequate test times can be planned. The final two major contributors are the goal reliability and test length, which directly influence cost and schedule. For this reason, various reliability growth strategies should be explored in order to optimize the reliability growth; balancing technical risk with program schedule and budget. Formulating effective reliability growth strategies also requires that a vigorous root cause analysis and corrective action (RCA/CA) process be developed. Simply replacing failed components during the RGT; without removing the underlying failure mode, will not result in reliability growth. Understanding the root cause of surfaced failure modes and devising effective mitigation strategies is the only method that will achieve reliability growth. This is why total calendar time that includes actual test time and time to perform RCA/CA must be considered and planned into the RGT and program test schedule. Another aspect to consider in planning the RGT is to identify items that are not likely to undergo reliability growth. The use of Commercial-Off-The-Shelf (COTS) items in many DoD programs is a reality that cannot be ignored. These COTS items will have no effective growth avenues since COTS suppliers are unlikely to change product designs. The RGT can be used to assess COTS ability to meet allocated reliability requirements and compliance with product specifications and warranties. The RGT can also provide an opportunity to assess the success of the COTS design integration process and robustness in the actual use environment.