Risk Informed Design modeling process & design team - Analyst interaction

As demand for highly reliable complex systems increases, engineers are being forced to consider the risk implications of design decisions earlier in the conceptual phase of projects and with greater accuracy. Standard probabilistic risk assessments (PRA) usually employed to verify that a product meets requirements are too resource intensive and too slow to keep up with the speed at which the design is maturing; while classical qualitative methods do not provide the level of detail and granularity required by the designers to make high-quality risk informed decisions. The Altair design team was able to overcome these challenges by employing a process of Risk Informed Design utilizing the Valador Reliability Tool (VRT). This tool is able to quickly and accurately produce estimates of the risk of Loss of Mission (LOM) and Loss of Crew (LOC) per mission and provide insight to the designers as to how their decisions will impact overall mission success. The VRT employs a method or risk assessment that is unlike traditional PRA as it effectively engages the designers in the model building process and as a result of this increased Designer Analyst interaction both the quality of the design and the quality of the PRA model is increased. This method of PRA seeks to create a baseline model by first capturing a complete set of initiating failure events which can lead to LOM/LOC based upon the Master Equipment List (MEL), dynamic mission events and identified hazards. Next, the event trees of these failures are generated automatically by correctly identifying the response of the system to an initiating failure and the risk exposure times of each failure mode through the use of schematics, designer interviews, and a priori knowledge. Once a baseline configuration has been captured in the reliability model, the tool facilitates the isolation of specific subsets of risk for directed trade studies. Now, the flexibility and speed of the tool can be leveraged to rapidly produce a large number of design options guided by the initial LOM/LOC scoring of the subset of components. The VRT gives the analyst the capability to model, score and analyze options in real-time with the designer. This provides immediate feedback and allows for a rapid iterative process which gives the team more freedom to effectively search the risk design space and find the "differences that make a difference." Ultimately, a scatter plot of LOM/LOC vs. Delta-Mass can be used to compare the relative implications of each option to one another by highlighting the cost and benefit of each option and the overall trend of the curve. Furthermore, the design team is able to compare options across trade study boundary lines, enabling a global risk perspective. The insight into the risk reduction efficiency of each option aides the design team in determining the best ways to spend mass to increase reliability across the vehicle given a constrained mass budget. Moreover, the VRT serves as an excellent method for vehicle integration to capture cross system dependencies, to find errors in the MEL and schematics, and to identify synergistic relationships that may not be immediately apparent due to the 'stove piping' of the design into specific subsystems.