L13 - Quarterbacking a Quality Risk Decision - Wednesday 10:00 am CDT
Trish Borzon
1370 Posts
Wednesday, May 26th at 10:00 am CDT
This presentation is meant to present a case study for using a “Quarterbacking” approach to technical problem solving. The strategy can be colloquially defined through an analogy to the Quarterback in American Football. A football quarterback has a limited number of options available to chose from, which are limited by the established rules of the game and the personnel defined by the play call (strategy). Similarly, the quality professional who is placed in the role of quarterback does not have control over the resources presented to them. Quarterback success is realized by a similar type of agile decision making which, particularly when based on Pareto ranking, resembles a “check down” strategy employed in football to manage risks and actualize gains. The “check down” strategy is utilized by the quarterback to define a sequential action plan based upon the potential reward (upside) for an action as well as the probability of success for that action. In order to maximize the potential reward, the quarterback monitors 3-5 potential actions which are developing in parallel. The first option, the down-field pass, presents the highest potential gain but faces the most obstacles. The “check down” option has less assured gain but also has few obstacles to success. There are also intermediate options that have more balanced risk and reward. In the quality corollary, the reward is represented by the likelihood and affectability measures while the obstacles are presented by the measurability of the phenomenon. The likelihood measure determines the probability, based upon application knowledge, that the mechanism under study represents the primary root cause. The affectability score is concerned with the amount of latitude available to reparameterized or redesign the features associated with that potential cause. The measurability metric represents the complexity of tools required to observe, measure, and quantify the effects of a potential root cause. This will be an in-depth case study of an engineering solution where multiple design features provide strength and a multitude of energies exist in the application to cause damage. Pareto ranking based upon the product domain knowledge will be used to construct the “check down” list for investigating the potential root causes as well as the potential design remedies. Each item on the list will be evaluated on the three measures defined above: Likelihood, affectability, and measurability. Three to five potential causes will be evaluated in parallel with emphasis on the most likely root cause, regardless of difficulty, as well as most easily defined potential cause, regardless of criticality. This sorting procedure is meant to increase the speed in determining the primary root cause and implementing corrective solutions to increase field reliability. While this prioritization strategy will improve decision making cycle times, one must also consider how to accelerate the assessments that drive these decisions. The author believes that the check down strategy is most effective when paired with simple tools that can quickly assess the effect of potential causes and the efficacy of potential fixes. These simple tools provide empirical evidence to establish a critical path for design and production team. There comes a critical watershed moment once the critical path is defined where the quarterback must transition to more sophisticated tools that can provide quantitative data that can be used for reliability statistics. This process will be demonstrated as a case study involving a refrigeration compressor that is used in intermodal (land, sea, and rail) transportation applications. The primary field observation was a compressor that was fully operational and met efficiency standards but was creating a large amount of noise. It was observed that the compressor did not become noisy until it was in the field for several years. Identifying and correcting the root cause required investigation to define: *The components, assemblies, and systems that represent the source of the observed noise. *Simple, non-destructive tools to identify noisy compressors in situ. *Simple, non-destructive tools to measure degradation of the affected components during accelerated testing. *The application energy or energies that contributed to the noisy operation. *The design parameters that could be augmented to increase the design strength or mitigate the transmission of energy into the affected components. *Simple analytical tools to qualitatively assess the best design option. *Robust quantitative tools that connect the analytical models, accelerated tests, and field energy measurement to perform reliability calculations and predict the reliability growth as design changes are implemented. While all phases of the case study will be presented, the quarterbacking role was most crucial in stages 2-6. In order to successfully implement a quarterback strategy, you must find simple ways to measure the degradation or change of components over time. Only after building a defined toolbox can you effectively investigate the causes and solutions to the observed failure modes. While the final stages of reliability growth demonstration are still underway, this team utilized a the Applied Mechanics knowledge management system to retain all of the data at each step of this case study such that different tests and tools could be combined to make accurate predictions regarding the improvement in field reliability. Further, this was critical to documenting this learning process so that it could be more quickly applied in other applications.