Sequential Sampling With Kernel-Based Bayesian Network Classifiers

2011 ◽  
Author(s):  
David Shahan ◽  
Carolyn C. Seepersad
Keyword(s):  
Bayesian Network ◽  
Collaborative Design ◽  
Design Space ◽  
Sequential Sampling ◽  
Space Filling ◽  
Training Set ◽  
Design Problems ◽  
Sampling Schemes ◽  
Domain Specific ◽  
Bayesian Network Classifiers

Complex design problems are typically decomposed into smaller design problems that are solved by domain-specific experts who must then coordinate their solutions into a satisfactory system-wide solution. In set-based collaborative design, collaborating engineers coordinate themselves by communicating multiple design alternatives at each step of the design process. Previous research has demonstrated that classifiers can be a communication medium for facilitating set-based collaborative design because of their ability to divide a design space into satisfactory and unsatisfactory regions. The proposed kernel-based Bayesian network (KBN) classifier uses a set of example designs of known acceptability, called the training set, to create a map of the satisfactory region of the design space. However, previous implementations used deterministic space-filling sampling sequences to choose the training set of designs. The shortcoming of deterministic space-filling sampling schemes is that they do not adapt to focus the samples on regions of interest to the design team (exploitation) or, alternatively, on regions in which little information is known (exploration). In this paper, we introduce the use of KBN classifiers as the basis for sequential sampling strategies that can be exploitive, exploratory, or any combination thereof.

scholarly journals Bayesian Network Classifiers for Set-Based Collaborative Design

2012 ◽  
Vol 134 (7) ◽  
Author(s):  
David W. Shahan ◽  
Carolyn Conner Seepersad
Keyword(s):  
Bayesian Network ◽  
Collaborative Design ◽  
Design Problem ◽  
Design Space ◽  
Expert Knowledge ◽  
Optimization Techniques ◽  
Domain Specific ◽  
Bayesian Network Classifiers ◽  
New Information ◽  
Engineering Design Problems

Complex engineering design problems are often decomposed into a set of interdependent, distributed subproblems that are solved by domain-specific experts. These experts must resolve couplings between the subproblems and negotiate satisfactory, system-wide solutions. Set-based approaches help resolve these couplings by systematically mapping satisfactory regions of the design space for each subproblem and then intersecting those maps to identify mutually satisfactory system-wide solutions. In this paper, Bayesian network classifiers are introduced for mapping sets of promising designs, thereby classifying the design space into satisfactory and unsatisfactory regions. The approach is applied to two example problems—a spring design problem and a simplified, multilevel design problem for an unmanned aerial vehicle (UAV). The method is demonstrated to offer several advantages over competing techniques, including the ability to represent arbitrarily shaped and potentially disconnected regions of the design space and the ability to be updated straightforwardly as new information about the satisfactory design space is discovered. Although not demonstrated in this paper, it is also possible to interface the classifier with automated search and optimization techniques and to combine expert knowledge with the results of quantitative simulations when constructing the classifiers.

scholarly journals Bayesian Network Classifiers for Set-Based Collaborative Design

2010 ◽  
Author(s):  
David Shahan ◽  
Carolyn C. Seepersad
Keyword(s):  
Design Process ◽  
Collaborative Design ◽  
Probability Distributions ◽  
Design Problems ◽  
Fuzzy Membership Functions ◽  
Domain Specific ◽  
Bayesian Network Classifiers ◽  
Complex Design ◽  
Data Points ◽  
Aerial Vehicle

Complex design problems are typically decomposed into smaller design problems that are solved by domain-specific experts who must then coordinate their solutions into a satisfactory system-wide solution. In set-based collaborative design, collaborating engineers coordinate themselves by communicating multiple design alternatives at each step of the design process. The goal in set-based collaborative design is to spend additional resources exploring multiple options in the early stages of the design process, in exchange for less iteration in the latter stages, when iterative rework tends to be most expensive. Several methods have been proposed for representing sets of designs, including intervals, surrogate models, fuzzy membership functions, and probability distributions. In this paper, we introduce the use of Bayesian networks for capturing sets of promising designs, thereby classifying the design space into satisfactory and unsatisfactory regions. The method is compared to intervals in terms of its capacity to accurately classify satisfactory design regions as a function of the number of available data points. A simplified, multilevel design problem for an unmanned aerial vehicle is presented as the motivating example.

Bayesian Network Classifiers and Design Flexibility Metrics for Set-Based, Multiscale Design With Materials Design Applications

2014 ◽  
Author(s):  
Jordan Matthews ◽  
Timothy Klatt ◽  
Carolyn C. Seepersad ◽  
Michael Haberman ◽  
David Shahan
Keyword(s):  
Composite Materials ◽  
Bayesian Network ◽  
Design Space Exploration ◽  
Design Space ◽  
Effective Properties ◽  
Materials Design ◽  
System Level ◽  
Bayesian Network Classifiers ◽  
Multi Level ◽  
Level Design

A set-based approach is presented for solving multi-scale or multi-level design problems. The approach incorporates Bayesian network classifiers (BNC) for mapping design spaces at each level and flexibility metrics for intelligently narrowing the design space as the design process progresses. The approach is applied to a hierarchical composite materials design problem, specifically, the design of composite materials with macroscopic mechanical stiffness and loss properties surpassing those of conventional composites. This macroscopic performance is achieved by embedding small volume fractions of negative stiffness (NS) inclusions in a host material. To design these materials, the set-based, multilevel design approach is coupled with a hierarchical modeling strategy that spans several scales, from the behavior of microscale NS inclusions to the effective properties of a composite material containing those inclusions and finally to the macroscopic performance of components. The approach is shown to increase the efficiency of multi-level design space exploration, and it is particularly appropriate for top-down, performance-driven design, as opposed to bottom-up, trial-and-error modeling. The design space mappings also build intuitive knowledge of the problem and promising regions of the design space, such that it is almost trivial to identify designs that yield preferred system-level performance.

Hierarchical Design of Negative Stiffness Metamaterials Using a Bayesian Network Classifier1

2016 ◽  
Vol 138 (4) ◽  
Author(s):  
Jordan Matthews ◽  
Timothy Klatt ◽  
Clinton Morris ◽  
Carolyn C. Seepersad ◽  
Michael Haberman ◽  
...  
Keyword(s):  
Bayesian Network ◽  
Hierarchical Modeling ◽  
Effective Properties ◽  
Negative Stiffness ◽  
Mechanical Stiffness ◽  
Design Problems ◽  
Length Scales ◽  
Bayesian Network Classifiers ◽  
Design Variables ◽  
Modeling Level

A set-based approach is presented for exploring multilevel design problems. The approach is applied to design negative stiffness metamaterials with mechanical stiffness and loss properties that surpass those of conventional composites. Negative stiffness metamaterials derive their properties from their internal structure, specifically by embedding small volume fractions of negative stiffness inclusions in a continuous host material. Achieving high stiffness and loss from these materials by design involves managing complex interdependencies among design variables across a range of length scales. Hierarchical material models are created for length scales ranging from the structure of the microscale negative stiffness inclusions to the effective properties of mesoscale metamaterials to the performance of an illustrative macroscale component. Bayesian network classifiers (BNCs) are used to map promising regions of the design space at each hierarchical modeling level, and the maps are intersected to identify sets of multilevel solutions that are likely to provide desirable system performance. The approach is particularly appropriate for highly efficient, top-down, performance-driven, multilevel design, as opposed to bottom-up, trial-and-error multilevel modeling.

Bayesian Networks for Set-Based Collaborative Design

2009 ◽  
Author(s):  
David Shahan ◽  
Carolyn C. Seepersad
Keyword(s):  
Bayesian Networks ◽  
Design Process ◽  
Collaborative Design ◽  
Design Space ◽  
Expert Knowledge ◽  
Hill Climbing ◽  
Local Networks ◽  
Design Problems ◽  
Potential Benefits ◽  
Aerial Vehicle

A set-based approach to collaborative design is presented, in which Bayesian networks are used to represent promising regions of the design space. In collaborative design exploration, complex multilevel design problems are often decomposed into distributed subproblems that are linked by shared or coupled parameters. Collaborating designers often prefer conflicting values for these coupled parameters, resulting in incompatibilities that require substantial iteration to resolve, extending the design process lead time without guarantee of achieving a good design. In the proposed approach to collaborative design, each designer builds a locally developed Bayesian network that represents regions of interest in his design space. Then, these local networks are shared and combined with those of collaborating designers to promote more efficient local design space search that takes into account the interests of one’s collaborators. The proposed method has the potential to capture a designer’s preferences for arbitrarily shaped and potentially disconnected regions of the design space in order to identify compatible or conflicting preferences between collaborators and to facilitate a compromise if necessary. It also sets the stage for a flexible and concurrent design process with varying degrees of designer involvement that can support different designer strategies such as hill-climbing or region identification. The potential benefits are the capture of expert knowledge for future use as well as conflict identification and resolution. This paper presents an overview of the proposed method as well as an example implementation for the design of an unmanned aerial vehicle.

Bayesian Network Structure Optimization for Improved Design Space Mapping for Design Exploration With Materials Design Applications

2017 ◽  
Author(s):  
Conner Sharpe ◽  
Clinton Morris ◽  
Benjamin Goldsberry ◽  
Carolyn Conner Seepersad ◽  
Michael R. Haberman
Keyword(s):  
Machine Learning ◽  
Bayesian Network ◽  
Design Space ◽  
Computational Cost ◽  
Materials Design ◽  
Optimization Techniques ◽  
Machine Learning Algorithms ◽  
Data Set ◽  
Design Problems ◽  
Highly Nonlinear

Modern design problems present both opportunities and challenges, including multifunctionality, high dimensionality, highly nonlinear multimodal responses, and multiple levels or scales. These factors are particularly important in materials design problems and make it difficult for traditional optimization algorithms to search the space effectively, and designer intuition is often insufficient in problems of this complexity. Efficient machine learning algorithms can map complex design spaces to help designers quickly identify promising regions of the design space. In particular, Bayesian network classifiers (BNCs) have been demonstrated as effective tools for top-down design of complex multilevel problems. The most common instantiations of BNCs assume that all design variables are independent. This assumption reduces computational cost, but can limit accuracy especially in engineering problems with interacting factors. The ability to learn representative network structures from data could provide accurate maps of the design space with limited computational expense. Population-based stochastic optimization techniques such as genetic algorithms (GAs) are ideal for optimizing networks because they accommodate discrete, combinatorial, and multimodal problems. Our approach utilizes GAs to identify optimal networks based on limited training sets so that future test points can be classified as accurately and efficiently as possible. This method is first tested on a common machine learning data set, and then demonstrated on a sample design problem of a composite material subjected to a planar sound wave.

scholarly journals Developing Ideation Cards for Mixed Reality Game Design

2017 ◽  
Vol 3 (2) ◽  
Author(s):  
Richard Wetzel ◽  
Tom Rodden ◽  
Steve Benford
Keyword(s):  
Collaborative Design ◽  
Game Design ◽  
Design Space ◽  
Mixed Reality ◽  
Idea Generation ◽  
Physical World ◽  
Pervasive Games ◽  
Domain Specific ◽  
Iterative Development ◽  
Different Types

Mixed reality games (MRGs) encompass a variety of gaming genres such as pervasive games, location-based games, and augmented reality games. They enrich the physical world with technology to create new and exciting possibilities for games – but at the same time introduce new challenges. In order to make the vast design space of MRGs easily accessible we have developed our Mixed Reality Game Cards. These are a deck of ideation cards that synthesize design knowledge about MRGs and enable collaborative design in a playful manner. In this paper, we describe the iterative development of the Mixed Reality Game Cards over the course of six studies. The final version of the cards constitutes a helpful tool for future designers of MRGs both for rapid idea generation as well as for more in-depth idea development. We achieve this by utilizing different types of domain-specific cards (Opportunities, Questions, Challenges) as well as promoting the inclusion of domain-extrinsic Theme cards and suggesting different rules for interacting with the cards.

A Set-Based Method for Multilevel Design of Materials and Structures With Manufacturing Uncertainty

2016 ◽  
Author(s):  
Clinton Morris ◽  
Carolyn C. Seepersad ◽  
Michael R. Haberman
Keyword(s):  
High Performance ◽  
Design Space ◽  
Probability Distributions ◽  
Joint Probability ◽  
Mechanical Stiffness ◽  
Design Problems ◽  
Paper Manufacturing ◽  
Bayesian Network Classifiers ◽  
Multi Level ◽  
Joint Probability Distributions

Recent research has indicated that embedding small volume fractions of negative stiffness (NS) inclusions within a host material can create composites with macroscopic mechanical stiffness and loss properties that exceed conventional composites. To design these composites, a multi-level, set-based approach that employs Bayesian network classifiers was developed to identify sets of satisfactory designs at each level of the multilevel design space. In this paper, manufacturing uncertainties are incorporated to further refine the design space mappings created by the set-based approach. Manufacturing uncertainty refers to the random deviations in dimensions and other properties that often arise when fabricating a specimen. Joint probability distributions are used to model this manufacturing uncertainty. The joint probability distributions are formulated as kernel density estimates that can be based on manufacturing data. The joint probability distributions are incorporated within the set-based approach to identify sets of designs that not only yield satisfactory performance but also offer robustness to manufacturing uncertainty. The approach is demonstrated in the context of hierarchical composite materials, but it can be applied to other multi-level design problems to efficiently yield sets of robustly manufacturable, high performance designs.

Sign in / Sign up

Export Citation Format

Share Document