Reliability growth program. 

Context in source publication

Context 1

... change in the U.S. Department of Defense reliability policy dictated by insufficient attention to reliability during product development will trigger some changes in program management as well as in the systems engineering organizations. That is why it is extremely important to capture positive lessons from successful programs such as the Stryker Mobile Gun System (MGS). In this article, the authors discuss three major factors that ensured the MGS program met its reliability requirements during product verification testing (PVT): N Program Management–Integrated Team, N Systems Engineering–Reliability Attainment, N Reliability Growth Analysis. The main intent of this article is to illustrate practical applications of these factors and some near- term payoff programs should receive in terms of performance and reliability. The Stryker family of vehicles is an eight-wheeled military combat vehicle being used by the Stryker Brigade Combat Teams and assembled into 10 different variants with a common chassis (Figure 1) . Eight main designs were developed by General Dynamics Land Systems (GDLS) as the prime contractor, successfully tested, and then fielded with the U.S. Army during 2003–2005. The Stryker MGS is by far the most complex and heaviest design of all the variants within the Stryker family (Figure 2) . It incorporates the common Stryker chassis and low profile turret with 105-mm gun that is equipped with an ammunition handling system and auto-loader. The Product Qualification Test (PQT) conducted in 2003 revealed a variety of reliability and performance issues within the MGS design, especially with the ammunition handling system and the mission equipment package. Between 2003 and 2006, program management made unprecedented efforts to redesign the MGS mission equipment package with an emphasis on its ammunition handling system. GDLS took the chal- lenge and dramatically revitalized its systems engineering organization. Such efforts set the stage for an increase in reliability during the redesign stage and then use of the proper Test-Find-Fix-Test procedure during PVT. The first reliability growth plan devel- oped by a group of internal and external reliability experts established a planned reliability growth curve that connected an engineering process with measured reliability. Interestingly, predicted reliability for PVT was very close to the actual demonstrated reliability in 2008. There are two main stages of product development in any program design or redesign activities and reliability growth testing. In order to achieve reliability requirements during design and subsequent test stages, the engineering community must employ robust engineering principles during the design stage and then manage failure modes during the test stage with a wide scope of timely issued corrective actions. Thus, the systems engineering team ensures initial reliability growth and then continues to develop improvements during the test phase. The program management team provides detailed schedule, proper budget, and resource management that supports the engineering team. And finally, the interpretation of the data from the test using reliability data analysis will direct the engineering efforts and will provide a proper assessment of the existing and/or potential reliability of the system. Below we will discuss all three elements in greater detail. An initial assessment of Stryker MGS reliability during PQT revealed the shortcomings of the existing reliability growth program. The program management team developed the following plan to address the reliability issues: Phase I—Additional reliability testing to evaluate effectiveness of the corrective actions developed from PQT, N Phase II—Systems engineering process improvement, N Phase III—Redesign of major subsystems and integration. These phases took place between 2003 and 2006 and then the program went into PVT in 2006. The main emphasis during these steps was made on systems engineering revitalization that will be discussed in the next section of this article. A Systems Engineering Reliability Growth Plan was developed to include both redesign activities and planned reliability growth testing. It is important to point out that during the design or redesign stage of the reliability growth program (Figure 3) the engineering team focused on an inherent reliability or hardware/software reliability. The main efforts of the design process target the ability of the system design to perform its function reliably and robustly over a useful lifetime. On the other hand, the next phase of the Reliability Growth Plan will uncover problems affecting the operational reliability, i.e., inherent and induced failures. The latter can be described as operator/user errors, maintenance errors, accidents, etc. We will discuss those categories of failures later in this article. The same systems engineering process described here can address both aspects of operational reliability during both phases. The program management team, working together between the Program Management Office Stryker Brigade Combat Teams and GDLS, were able to plan, budget, and execute the Reliability Growth Plan successfully. Root cause analysis process followed by verification and validation of the corrective actions process became the major driving force behind the reliability growth of the MGS. Communication and explicit information about design deficiencies, verified fixes, and validation processes were key contributors to the overall success of the program. Engineering information about system performance during testing can be considered as feedback of the process that had designed such a system. It became obvious that current SE processes lacked focus on the reliability of the system. This conclusion triggered a systems engineering revitalization process that had system reliability as a main deliverable of the SE process. In addition to a very well defined SE master plan that served as guidance for the MGS redesign processes, the SE organization must have solid processes that govern every day activities, and SE management must have the associated metrics that adequately measure such processes. Thus, the SE organization focused on reliability processes, and appropriate management metrics formed the engineering core that was instrumental in achieving reliability requirements. With the help of an external consultant, a revitalized SE process was developed and later used with great success on the MGS program. The process combines analysis and review of the system reliability requirements, system and subsystem design (redesign) for reliability, testing for reliability, and corrective actions tracking. A multifunctional and multilevel team of system and subsystem engineers formed a Failure Prevention and Review Board that became the driving force of the design improvement and was led by the Program Management Office. Such a process was developed and copyrighted by Dr. L. Crow and is presented in Figure 4 . The Design Actions Reporting and Tracking (DART) process discussed here manages the discov- ered failure modes as well as associated corrective actions through a redesign process driven by the Failure Prevention Review Board. Each DART created for an individual failure mode by an Incident Screening Team defines the seed of the database that can be used as a management measure of the process. Thus, we have all elements of the successful process—the multifunctional engineering organization, a well defined process, and management metrics to adequately assess both the flow and aging of the process. Also, it was found extremely useful to form affinity teams that address different common aspects of the design, such as a fasteners team, leak prevention team, integration team, etc. Because of the length limitations of this article, it is impossible to describe all the important steps, elements, and milestones of the GDLS SE process. However, a few extremely important elements must be noted. The DART process generates a closed-loop failure mitigation system that not only drives the engineering correction process, but also helps to make statistical inferences from the test. Furthermore, the DART process or any other Failure Reporting and Corrective Action System connected to a Design Failure Mode and Effect Analysis or Failure Mode, Effect, and Criticality Analysis as a failure mode discovery mechanism can be the main driving force of the design for reliability approach. This methodology is being used by GDLS now on other programs. It is imperative to note that major elements of the SE process initiated on the MGS program are described in the new ‘‘Reliability Program Standard for Systems Design, Development and Manufactur- ing.’’ 3 It summarizes the four main objectives of the new standard: N understand the requirements, N design for reliability, N produce reliable system, N field and maintain the product. The first three objectives correlate to the described above DART process. The last factor of a successful program is reliability data analysis. Indeed, the final reliability test is ultimately feedback on the previously described processes. Without proper inferences derived from the test and adequate data analysis, it is impossible to measure the ...