Marin D. Guenov

A methodology for modelling manufacturing costs at conceptual design

This paper describes a method for modelling costs throughout the design phase of a products life-cycle, from conceptual to detail design. The timely provision of cost data facilitates the quantitative evaluation of designs through a cost function. This approach to design evaluation has the advantages of allowing management to make more accurate bid estimates, encouraging designers to design to cost and reducing the amount of design rework, hence reducing the products time to market and controlling product cost. The cost modelling strategy adopted incorporates the use of knowledge-based and case-based approaches. Cost estimation is automated by linking design knowledge, required for predicting design features from incomplete design descriptions, to production knowledge. The link between the two knowledge paradigms is achieved through the blackboard framework of problem solving, which incorporates both case-based and rule-based reasoning. The method described is aimed at innovative design activity.

Graph-based feature recognition for injection moulding based on a mid-surface approach

This paper presents a novel CAD feature recognition approach for thin-walled injection moulded and cast parts in which moulding features are recognised from a mid-surface abstraction of the part geometry. The motivation for the research has been to develop techniques to help designers of moulded parts to incorporate manufacturing considerations into their designs early in the design process. The main contribution of the research has been the development of an attributed mid-surface adjacency graph to represent the mid-surface topology and geometry, and a feature recognition methodology for moulding features. The conclusion of the research is that the mid-surface representation provides a better basis for feature recognition for moulded parts than a B-REP solid model. A demonstrator that is able to identify ribs, buttresses, bosses, holes and wall junctions has been developed using C++, with data exchange to the CAD system implemented using ISO 10303 STEP. The demonstrator uses a commercial algorithm (I-DEAS) to create the mid-surface representation, but the feature recognition approach is generic and could be applied to any mid-surface abstraction. The software has been tested on a range of simple moulded parts and found to give good results.

A methodology for knowledge management implementation

This article describes research work which was directed towards providing the automotive industry with a practical methodology that translates the conceptual ideas of knowledge management (KM) into a working programme with defined objectives, using industry terminology. The research also developed a supporting analysis methodology that enables an effective analysis of the influences on employee activities when creating and sharing valuable corporate knowledge, that spans technical and cultural boundaries. This happens through identifying the factors that impact on defined KM metrics. The analysis identifies the key influencing factors within a working environment. The research benefits are felt when the ground‐level drivers of KM behaviour are improved through links to an appropriate KM strategy. KM strategy may emphasise organisational cultural changes or IT changes or both in an endeavour to improve innovation, reduce business costs and reduce time to market of new products. An industrial case study was undertaken to validate the research.

Novel uncertainty propagation method for robust aerodynamic design

Starting from a comparative study of various methods for uncertainty propagation, this paper presents a novel reduced quadrature technique to be used in gradient-based robust design optimization of aerodynamic shapes. The accuracy and computational efficiency of the method are investigated by means of mathematical analyses and numerical examples. The method is then applied to the robust design of airfoils under probabilistic uncertainty. It is shown that the solutions obtained through the proposed method can outperform those obtained through linearization, without any significant increase in computational cost.

A Comparison of Airfoil Shape Parameterization Techniques for Early Design Optimization

Parameterization has a significant role in airfoil design and optimization where, especially in the early design stages, the aim is to be able to explore as many design alternatives as possible while keeping the number of design parameters to a minimum. Several desirable characteristics, namely, Parsimony, Intuitiveness, Orthogonality, Completeness and Flawlessness are considered. These are applied as comparison metrics to five airfoil parameterization techniques: Ferguson’s curves, HicksHenne bump functions, B-Splines, PARSEC, and Class/Shape function Transformation. The method allows each parameterization technique to be ranked according to each criterion considered, thus providing a basis for objective multiple criteria decision making.

Robust Aircraft Conceptual Design Using Automatic Differentiation in Matlab

The need for robust optimisation in aircraft conceptual design, for which the design parameters are assumed stochastic, is introduced. We highlight two approaches, first-order method of moments and Sigma-Point reduced quadrature, to estimate the mean and variance of the design’s outputs. The method of moments requires the design model’s differentiation and here, since the model is implemented in Matlab, is performed using the automatic differentiation (AD) tool MAD. Gradient-based constrained optimisation of the stochastic model is shown to be more efficient using AD-obtained gradients than finite-differencing. A post-optimality analysis, performed using AD-enabled third-order method of moments and Monte-Carlo analysis, confirms the attractiveness of the Sigma-Point technique for uncertainty propagation.

Airfoil Design under Uncertainty with Robust Geometric Parameterization

[Abstract] The subject of this paper is the conceptual design of airfoils under geometric uncertainty. In this context a novel methodology is presented, which couples an innovative uncertainty propagation technique with a hybrid optimization method to alleviate the computational burden of optimization under uncertainty. In such an approach, a key role is played by the adopted geometric parameterization, which determines significant features of the design space, hence influencing the optimization process. Identifying the regions of the design space in which erroneous profiles are more likely to appear can be beneficial for the optimization. For this purpose, a method for localizing flawed regions in the design space has been developed and is presented in this paper as part of the novel airfoil design methodology. The optimization strategy under uncertainty and the robust geometric parameterization techniques are evaluated on a test case making use of the viscous 2D computational code VGK and the PARSEC-11 parameterization. The results demonstrate a significant reduction of the number of failures due to erroneous parameterization occurring during the optimization process while the hybrid optimizer proved capable of finding meaningful robust solutions in an efficient way. Although the methodology was demonstrated on an airfoil parametric optimization, it is emphasized that the methods proposed in this paper are fairly generic and can be applied to a broad range of design optimization problems of multidisciplinary nature.

Similarity measures for mid-surface quality evaluation

Mid-surface models are widely used in engineering analysis to simplify the analysis of thin-walled parts, but it can be difficult to ensure that the mid-surface model is representative of the solid part from which it was generated. This paper proposes two similarity measures that can be used to evaluate the quality of a mid-surface model by comparing it to a solid model of the same part. Two similarity measures are proposed; firstly a geometric similarity evaluation technique based on the Hausdorff distance and secondly a topological similarity evaluation method which uses geometry graph attributes as the basis for comparison. Both measures are able to provide local and global similarity evaluation for the models. The proposed methods have been implemented in a software demonstrator and tested on a selection of representative models. They have been found to be effective for identifying geometric and topological errors in mid-surface models and are applicable to a wide range of practical thin-walled designs.

Requirements-Driven Design Decomposition: A Method for Exploring Complex System Architecture

A design decomposition model is proposed in which Axiomatic Design matrices (DM) map the Functional Requirements (FRs)-Design Parameters (DPs) relationship, and Design Structure Matrices (DSM) model the system development context. In the proposed model the DM and the DSM co-evolve. This approach is better suited for industries where the new product development process is constrained by the need to utilise existing manufacturing processes, to accommodate limitations of the supply chain and so forth. Traversing between the two types of matrices allows the application in a controlled manner of the system knowledge which surrounds the decision making process and the definition of the system architecture. A case study, conducted as part of our research, has verified the need for such a model. The study demonstrated also the need to understand more clearly the effect of constraints, and their associated project risk. This need is reflected in the development of a prototype software tool, the idea for which is described in this paper.Copyright

Covariance structural models of the relationship between the design and customer domains

Abstract This paper addresses the problem of modelling and mapping of ‘difficult to quantify’ customer needs to technical requirements and subsequently to design parameters. Proposed is a covariance structural equation model, which incorporates a confirmatory and a structural component. The former is used for the decomposition of the qualitative customer needs, modelled as latent variables, onto a generally larger number of measurable technical requirements. The structural component maps the technical requirements to design parameters. The concept is illustrated by an example. The model is confined to the linear dependence between the variables, but in general the approach can handle a number of non-linear relations through variable transformation. The conclusion is that the proposed synthetic procedure, named Structural Equation Models for the Design of Engineering Systems, represents a sufficiently rich and generic structure capable of bridging the gap between the customer and the design domains.