Bryan M. O’Halloran

The Early Design Reliability Prediction Method

The purpose of this paper is to formalize the Early Design Reliability Prediction Method (EDRPM) into a comprehensive framework, then to provide a case study using an Electrical Power System (EPS) which shows the usefulness of the methods. EDRPM has been developed to facilitate decision making in early design using quantitative reliability results [1]. Candidate components and design alternative are eliminated using justification provided by EDRPM. The output of this method is a set of design alternative that have a reliability values at or greater than a preset reliability goal. At the completion of applying EDRPM, additional metrics can be used to determine a final design. This research addresses the need for reliability methods to be moved earlier in the design process. Current methods are applicable after components have been selected. EDRPM is used during functional design, and when concepts are generated. This method also calculates functional failure rates which are applied to generate the function and component distributions. The results of the case study shows that several candidate components and design alternatives can be eliminated using EDRPM. It is demonstrated that only a subset of designs that meet the failure rate piece of the reliability goal should not be eliminated. The reliability goal is the combination of two parts; the failure rate and the probability of not exceeding the failure rate. Several of these design still have a probability of exceeding the second piece of the reliability goal given that they meet the first.Copyright

Link Between Function-Flow Failure Rates and Failure Modes for Early Design Stage Reliability Analysis

This scope of this paper is to provide an extension to the Function Failure Design Method (FFDM). We first implement a more robust knowledge base using Failure Mode/Mechanism Distributions 1997 (FMD-97). Then failure rates from Nonelectric Parts Reliability Data (NPRD-95) are added to more effectively determine the likelihood that a failure mode will occur. The proposed Functional Failure Rate Design Method (FFRDM) uses functional inputs to effectively offer recommendations to mitigate failure modes that have a high likelihood of occurrence. This work uses a past example where FFDM and Failure Modes and Effects Analysis (FMEA) were compared to show that improvements have been made. A four step process is presented to show how the FFRDM is used during conceptual design.Copyright

A Method to Calculate Function and Component Failure Distributions Using a Hierarchical Bayesian Model and Frequency Weighting

This paper presents a method to calculate function and component parameter distributions during the design process. Frequency Weighting, a unique style of weighting proposed in this research, is applied to a Hierarchical Bayesian model to account for the number of times a component has solved a function. During the design process, functions are systematically solved by components to transition from a functional model to a physical design. This research contributes to an ongoing effort toward predicting reliability in early design, specifically during functional modeling and concept generation. In general, reliability prediction methods are applied after concept generation. There currently does not exist a statistical method to calculate functional failure rates to aid reliability prediction during and before concept generation. The method presented in this paper also captures uncertainty in the early stages of design. This is important because uncertainty in this stage of the design process can be significant. A description of the process used to calculate the function and component level failure rate distributions is presented. The level of detail provided is meant for reapplication to other examples. Three examples are worked out and graphical results are presented. These results show an effect of the Frequency Weighting on the function level distribution. Changing the occurrence vector, which is used to show the number of times a set of components has solved a function, from (1, 1, 1, 1) to (1, 1, 2, 5) results in the function level distribution mean value shifting from 5.53E−06 to 4.84E−06. In addition, an example is provided to demonstrate how this method can be applied while components are being selected during the design process. A two part reliability goal is generated for the combined failure rate of the design and the probability a design will meet that goal. Function level distributions are used to show which components should initially be selected to maintain reliability values that meet the reliability goal. Combinations of compatible component level distributions are also used to calculate a combined failure rate distribution for each design. A probability is calculated for each distribution to show which designs meet the probability portion of the reliability goal.Copyright

Early Design Stage Reliability Analysis Using Function-Flow Failure Rates

In this paper, failure rates for function-flow pairs are presented. This data creates an opportunity for the designer to move reliability analysis earlier in the design process. The function-flow failure rates can be used to make design decisions before components are selected giving the designer increased knowledge to explore alternative options. A reliability block diagram approach has been adopted to evaluate the reliability of three designs at both the functional and component level. The results show that the bounds from the functional reliability overlap those of the component reliability.Copyright

Simulation Based Machine Learning for Fault Detection in Complex Systems Using the Functional Failure Identification and Propagation Framework

Fault detection and identification in mechatronic systems with complex interdependencies between subsystems is a very active research area. Various alternative quantitative and qualitative methods have been proposed in the literature for fault identification on industrial processes, making it difficult for researchers and industrial practitioners to choose a method for their application. The Functional Failure Identification and Propagation (FFIP) framework has been proposed in past research for risk assessment of early complex system designs. FFIP is a versatile framework which has been extended in prior work to automatically evaluate sets of alternative system designs, perform sensitivity analysis, and event trees generation from critical event scenario simulation results. This paper’s contribution is an FFIP extension, used to generate the training and testing data sets needed to develop fault detection systems based on data driven machine learning methods. The methodology is illustrated with a case study of a generic nuclear power plant where a fault or the location of a fault within the system is identified. Two fault detection methods are compared, based on an artificial neural network and a decision tree. The case study results show that the decision tree was more meaningful as a model and had better detection accuracy (97% success in identification of fault location).Copyright

Applying Designer Feedback to Generate Requirements for an Intuitive Biologically Inspired Design Tool

This research provides a comparison of a set of Bio-Inspired Design (BID) tools to determine advantages and disadvantages of each, with particular focus on database-directed design processes. The result of this comparison is a set of beneficial attributes, discussed to develop requirements for formulating an effective BID tool. In this study, each tool is evaluated in terms of both the effectiveness of the concept generation process, and designer feedback concerning the effective elements. The comparison between tools uses concept sketches and feedback generated from a classroom of graduate and undergraduate engineering students. Over the course of a ten-week class, each BID technique was formally presented to the students. Following this presentation, students were given a new design problem and instructed to use the new BID technique to generate a set of solution concepts. Quantity of concepts generated was used to assess the goodness of each concept generation activity outcome, which forms one basis for comparing the different tools. In addition, questionnaires were used to assess and identify the various positive and negative elements of each BID tool.Copyright

A Plant-Wide and Function-Specific Hierarchical Functional Fault Detection and Identification (HFFDI) System for Multiple Fault Scenarios on Complex Systems

The development of Fault Detection and Identification (FDI) systems for complex mechatronic systems is a challenging process. Many quantitative and qualitative fault detection methods have been proposed in past literature. Few methods address multiple faults, instead an emphasis is placed on accurately proving a single fault exists. The omission of multiple faults regulates the capability of most fault detection methods. The Functional Failure Identification and Propagation (FFIP) framework has been utilized in past research for various applications related to fault propagation in complex systems. In this paper a Hierarchical Functional Fault Detection and Identification (HFFDI) system is proposed. The development of the HFFDI system is based on machine learning techniques, commonly used as a basis for FDI systems, and the functional system decomposition of the FFIP framework. The HFFDI is composed of a plant-wide FDI system and function-specific FDI systems. The HFFDI aims at fault identification in multiple fault scenarios using single fault data sets, when faults happen in different system functions. The methodology is applied to a case study of a generic nuclear power plant with 17 system functions. Compared with a plant-wide FDI system, in multiple fault scenarios the HFFDI gave better results for identifying one fault and also was able to identify more than one faults. The case study results show that in two fault scenarios the HFFDI was able to identify one of the faults with 79% accuracy and both faults with 13% accuracy. In three fault scenarios the HFFDI was able to identify one of the faults with 69% accuracy, two faults with 22% accuracy and all three faults with 1% accuracy.Copyright

A Method to Compute Early Design Risk Using Customer Importance and Function-Flow Failure Rates

The general method for using customer importance to validate a design misses the important opportunity to influence how that design is created. The intent of this research is to capture risk during the conceptual design stage. Risk is calculated using function-flow failure rates and customer importance. This allows the designer to effectively identify what functionality should be given additional importance during the generation and selection of design concepts. Functional risk using customer importance has not yet been investigated. In general, risk is implemented later in the design process. A generic process to calculate the risk is presented, then applied to an example where a subset of function-flows have been identified as generating 75 % of the risk.

Developing New Design Requirements to Reduce Failures in Early Complex Systems Design

In this paper, we develop a method that produces recommendations, usable by a designer, that reduce the likelihood of a failure occurring. Prior work introduced the Function Failure Rate Design Method (FFRDM) which uses historical data as evidence to generate new design requirements. This paper presents improvements to FFRDM by including an iterative loop within the method that begins with specific recommendations. This allows evidence from the analysis to support the addition of new requirements and functionality into the design. Once the iterative loop has converged with no new requirements left to generate, all recommendations are used for concept generation. In addition, metrics are developed that can be used later to analyze the design. These metrics are important to ensure that the design has considered the full set of recommendations. Specifically, the updated FFRDM improves the original FFRDM with 1) a systematic and repeatable heuristic to group failure modes and determine which failure modes are predominate, 2) a categorization of the recommendations, 3) metrics built for the recommendations used in concept generation to make them measurable, and 4) using recommendations to develop new requirements and functionality. To show the usefulness of each improvement to FFRDM, a case study using an electrical power system (EPS) is provided.Copyright

A Survey of Early Design Risk and Reliability Methods and Their Impediments to Move Into Practice

This research surveys early design risk and reliability methodologies and discusses the impediments of moving these research methods into practice. Reliability engineering techniques exist primarily to help engineers better meet the needs of customers by extending design life and reducing the number of failures observed throughout the intended life. These efforts look at system components and functions, critical events, failure modes, and system characteristics to assess risk and reliability during the early design phase before detailed design has begun. Surveying early design reliability to identify underdeveloped areas of research contributes to an ongoing effort to increase the presence of reliability engineering earlier in design. In addition, this improves a researchers’ understanding of key consideration that need to be addressed during the development of the research so that it is useful in practice. Throughout this paper, four fundamental methods are identified and described including Event Tree Analysis, Fault Tree Analysis, Reliability Block Diagrams, and Failure Modes and Effects Analysis. Related methods, or those developed to solve limitations of the fundamental methods, are presented and compared to the fundamental methods. Finally, the impediments of moving research methods into practice are surveyed, then discussion is provided for the factors that improve this transition of research.Copyright