Ryan S. Hutcheson

Function Design Framework (FDF): Integrated Process and Function Modeling for Complex Systems

Functional models are representations of the energy, material and signal transformations that occur through the expected or normal operating condition of a product. As the complexity of products increases, there are often multiple dimensions to their operation in addition to their nominal operating state, e.g., crash protection systems in a car or laser leveling and stud finding combined in a single tool. Here system state is used to represent the different operational dimensions of a product, and a representation scheme that allows designers to fully explore system functionality of products with multiple system states is explored. Previous work in process and functional analysis is integrated to better represent complex systems with multi-dimensional system functionality. Process and functional modeling are integrated to produce a new function design framework supporting user-defined fidelity of hierarchical models for functional representation. An example modeling a complete automobile life cycle illustrates the development of integrated process and functional models within a complex system analysis.Copyright

Process and event modelling for conceptual design

Abstractions perform a fundamental role during product design, freeing a problem from reality into a representation more readily represented with engineering principles. Functional modelling provides such a representation for product design where customer needs are translated into a representation of elementary operations defining what a product must do to achieve a desired goal. With solely the generation of a functional model, the designer, however, runs the risk of failing to explicitly capture expected interactions and operations of the product as a whole. To that end, this paper presents a process modelling methodology consisting of model levels for the representation of product-related events and configurations based on current functional modelling techniques. Being based on functional modelling allows process modelling to integrate with functional modelling during conceptual design activities. The levels for the process model then collectively define customer needs related to how a product will be used, environments where a product will operate and changes that a product must undergo to meet customer expectations. To demonstrate the generation of event and configuration models, a common household product is investigated; this is followed by a case study discussion where process modelling is applied during the design of two ground robots.

A FUNCTION-BASED METHODOLOGY FOR ANALYZING CRITICAL EVENTS

The objective of this research was to develop a function-based method for analyzing the critical sequences of events that must occur for complex space missions to be successful. The resulting methodology, the Function-based Analysis of Critical Events, or FACE, uses functional and event models to identify the changes in functionality of a system as it transitions between critical mission events. Two examples are presented that detail the application of FACE to prior mission failures including the loss of the Columbia orbiter and the failure of the Mars Polar Lander probe. The result of the research is a methodology that allows designers to not only reduce the occurrence of such failures but also analyze the specific functional causes of the failures when they do occur.Copyright

APPLICATION OF A GENETIC ALGORITHM TO CONCEPT VARIANT SELECTION

This paper outlines a framework for applying a genetic algorithm to the selection of component variants between the conceptual and detailed design stages of product development. A genetic algorithm (GA) is defined for the problem and an example is presented that demonstrates its application and usefulness. Functional modeling techniques are used to formulate the design problem and generate the chromosomes that are evaluated with the algorithm. In the presented example, suitable GA parameters and the break-even point where the GA surpassed an enumerated search of the same solution space were found. Recommend uses of the GA along with limitations of the method and future work are presented as well.Copyright

Function-Based Behavioral Modeling

Behavioral models, mathematical models of a system’s ability to meet customer needs, are useful evaluation tools throughout the design process of systems. Currently, behavioral modeling is conducted at a component level. The models used to evaluate a system are associated exclusively with the components used to solve a product’s desired functionality. As a result, it is often difficult to create behavioral models during the early stages of design when these component solutions have not been identified. However, during these early stages of design, information about the desired functionality of the system is known. The objective of the work presented in this paper is to develop a method that uses this information, in the form of functional models, as the basis for creating behavioral models. The paper proposes a five step method for creating the behavioral models from the functional model. Significant contributions from the work include reuse of behavioral model elements based on common functionality, swapping of model elements with varying fidelity, a framework for mathematical concept evaluation and selection and the linking of assumptions made during mathematical modeling to their effects on the functionality of the product and vice-versa. Two examples of the method are included, a summary example of a resistor network and a complete example based on the dynamic modeling of a Formula SAE racecar. Conclusions from the work and examples are presented along with areas of future research.Copyright

A Hybrid Sensitivity Analysis for Use in Early Design

Sensitivity analyses are frequently used during the design of engineering systems to qualify and quantify the effect of parametric variation in the performance of a system. Two primary types of sensitivity analyses are generally used: local and global. Local analyses, generally involving derivative-based measures, have a significantly lower computational burden than global analyses but only provide measures of sensitivity around a nominal point. Global analyses, generally performed with a Monte Carlo sampling approach, and variation-based measures provide a complete description of sensitivity but. incur a large computational burden and require information regarding the distributions of the design parameters in a concept. Local analyses are generally suited to the early stages of design when parametric information is limited, and a large number of concepts must be evaluated (necessitating a light computational burden). Global analyses are more suited to the later stages of design when more information about parametric distributions is available and fewer concepts are under consideration. Current derivative-based local approaches provide a different and incompatible set of measures than a global variation-based analysis. This makes a direct comparison of local to global measures ill posed. To reconcile local and global sensitivity analyses, a hybrid local variation-based sensitivity (HyVar) approach is presented. This approach has a similar computational burden to a local approach but produces measures or percentage contributions. The HyVar approach is directly comparable to global variation-based approaches. In this paper, the HyVar sensitivity analysis method is developed in the context of a functional based behavioral modeling framework. An example application of the method is presented along with a summary of results produced from a more comprehensive example.

Applying Functional Modeling as a Unifying Basis for Design for Six Sigma Execution

This paper explores the applicability of the most recently developed methods in functional modeling to Design for Six Sigma transfer function development and requirements flowdown. An example created during a collaborative research project between the General Motors R&D Center and the University of Missouri – Rolla is used to demonstrate the benefits of using standardized functional modeling during conceptual design. The proposed standard for creating the functional models is the Functional Basis. The Functional Basis is a list of function and flow terms that can be used to describe electro-mechanical systems. The example presented in this paper is based on the parking brake system of a passenger car. Module heuristics, function-based rules for partitioning systems, were used to define the sub-systems during the requirements flowdown example. The functional modeling techniques used in this example provide a standard method of capturing current engineering design knowledge while allowing additional knowledge to be discovered.Copyright

Sensitivity measures for use during conceptual design

This paper outlines a function-based method for generating and utilising design parameter sensitivity knowledge for use during the conceptual design phase of an engineering system. Of particular interest is an understanding of sensitivity across a range of parameter values instead of only at a single nominal point. To do this, functional models created using the functional basis are created. Behavioural models for each function in the model are then identified and a sensitivity analysis is performed on each model with respect to each design parameter. A new proposed measure is used along with a coefficient of variation to reason about product sensitivity. The framework is suitable for storage in a repository and then used to aid the allocation of design and modelling resources during the design of a system with similar functionality. The specific focus areas of this paper are the sensitivity parameters and methods required to generate the sensitivity information for storage. An extensive case study to human powered flashlights is presented.

EFFECT OF MODEL ELEMENT FIDELITY WITHIN A COMPLEX FUNCTION-BASED BEHAVIORAL MODEL

The Function-based Behavioral Modeling (FBBM) design tool was introduced in prior work as a means of using formal functional modeling as the foundation for creating detailed mathematical models of system behavior. The overall objective of this work is to create a framework for partitioning modeling efforts into functional elements and promoting model storage and re-use through the use of functional models. In prior work, the FBBM method was introduced to model the complete vehicle dynamics of a Formula SAE racecar, highlighting the representation of functionality and the development of behavioral models. The objective of the work presented in the current paper is to demonstrate the ability to incorporate models of varying fidelity within a function-based behavioral model of a complex system. Additionally, the impact of model fidelity on the model’s predictions is addressed. A previously developed model is used as a foundation for developing the necessary new models and illustrating the impact of model fidelity on performance predictions when selecting a tire during early design. The results illustrate that the FBBM framework allows models of varying fidelity to be quickly made and their effect on predicted performance to be measured in order to assist critical early design choices.

Function-based behavioural modelling

Using mathematical behaviour models to determine system performance is crucial to the design of complex systems. Nevertheless, during the early stages of design, it is often difficult to create behavioural models as component solutions have not been fully identified. What is known is information about the desired functionality of the system is known. Particularly for complex system design, functional-based representations are important during the early stages of design. The objective of the work presented in this paper is to develop a functional model-based method for creating behavioural models to facilitate early behaviour model creation, model reuse, and concept evaluation. Significant contributions of this work include establishing the method and the formal addition of quantitative flow modelling to existing qualitative functional modelling methods. This quantitative flow modelling method enables a natural and consistent extension of qualitative functional modelling to quantitative behaviour modelling. An extensive example application of the method based on the dynamic modelling of a Society of Automotive Engineers Formula competition racecar is included to illustrate and verify the method.