Ola S Wagersten

A Framework for Non-nominal Visualization and Perceived Quality Evaluation

Perceived Quality clusters different aspects that influence the customers perception of non-functional quality on a product that are perceive through senses. All together those aspects and the harmony between them reflect the producers ability to control product parameters and thereby also mirror the functional quality of the product. High Perceived Quality can not be added to the product at the end of the developing process. Project prerequisites, system solution, factory capability etc. are criterion to succeed. Therefore, it is important to be able to evaluate Perceived Quality early in the process when product system solutions and architecture are defined, although data maturity is low. This paper presents a comprehensive framework to manage and support evaluation of Perceived Quality aspects in a product development process. The framework is based on an industrial process in combination with recent research within the field. The framework focus on activities that can be performed at different stages in the developing process based on maturity of the CAD or styling data. For example, if the styling data is divided into different components by split-lines it has reached higher level of maturity then styling data that not has been divided. Consequently, the choice of applied method is based on data maturity, regardless phase in the developing process. The framework contain methods based on several different simulation and analysis techniques. Both design methods, Computer Aided Tolerancing and FEA based non-rigid variation simulation are represented in the framework.

Parameters Influencing the Perception of Geometrical Deviations in a Virtual Environment

On car exteriors, geometrical variation, stemming from the manufacturing process, can distort the visible relationships between parts, i.e. the split-lines. In the automotive industry, non-nominal visualization is often implemented as a part of early concept reviews. Visualizing the results from a Computer-Aided Tolerancing analysis in a virtual environment allows to evaluate what deviations will be visible to customers and whether they can have a negative influence on the perceived product quality. The geometry, colors and materials of parts make some split-lines more visually sensitive than others because on them, potential deviations are especially disturbing to the onlooker. Moreover, the visual sensitivity is influenced by the context of product interaction in terms of illumination, reflections and the observer’s viewing angle. These factors need to be taken into account when determining tolerance levels to ensure that an industrial design concept will look good once produced. This paper presents a study of the influence split-line geometry and several visualization parameters have on the ability of test persons to identify geometrical deviations in a number of computer generated images. Additionally, eye-tracking was performed to identify what areas of the split-lines the test persons looked at while performing the task. The results confirm that the included split-lines are not equally sensitive to variation. Further, it is shown that the visualization settings indeed have an influence on the ability to detect variation but that the most sensitive settings differ between combinations of deviations and split-lines.Copyright

Influence of rigid and non-rigid variation simulations when assessing perceived quality of split-lines

Non-nominal visualisation is used to evaluate how geometrical variation influences the appearance of split-lines between components on a car body. Depending on the simulation method, visualised variation can be represented in an unrealistic manner, which can affect the perception of the result. This paper presents a comparative study in which eye-tracking equipment has been used in order to explore whether the interpretation of variation differs between rigid- and non-rigid-based variation simulations during the evaluation of visualised models. Subjects from the automotive industry were asked to evaluate two virtual models where the same amount of variation was represented by rigid and non-rigid models. Eye tracking was used to record how the subjects performed the assessment. The result shows that there is a significant difference in how simulation results are interpreted using rigid versus non-rigid simulation.

Non-Rigid Behavior Prediction Based on Styling Data for Evaluation of Perceived Quality

In the automotive industry it is of great importance to manage the impact of part and assembly variation as it affects aesthetical aspects of the final Perceived Quality, such as the appearance of flush, parallelism and gaps in split-lines. By using a CAT (Computer Aided Tolerancing) software simulating tolerances, in combination with a high-end visualization tool, evaluations are made to judge a predicted outcome of the final Perceived Quality. When a new styling concept needs to be evaluated the data used for analysis and visualization will be less mature and the fact that several vehicle components are non-rigid will further complicate the realization of non-nominal visualization. The deformation behavior of non-rigid components is dependent on several parameters such as part shape, position of reference points, flanges, reinforcements etc. Such information is most likely not available when evaluating a new styling concept, represented only by a surface skin. This paper discusses the possibility to use this limited information, defined as styling data, as a basis when performing non-nominal visualization and evaluation of Perceived Quality based on deformable non-rigid components. The insufficient information is supplemented by collecting and comparing data from prior projects to enable non-rigid behavior prediction of components as such.© 2009 ASME

Towards Non-FEA-Based Deformation Methods for Evaluating Perceived Quality of Split-Lines

In the automotive industry, the evaluation of Perceived Quality of split-lines is strongly dependent on simulation and visualisation activities to analyse consequences of geometrical variation. A truthful representation of the part behaviour is therefore essential. Moreover, variation simulation of non-rigid parts is today performed by finite element analysis (FEA). FEA-based methods demand meshed models that correspond to the final engineering design to calculate the correct stiffness matrix. However, geometrical models in early phases have significantly lower level of detail. Approximate methods are therefore considered as options to better deal with this restriction. In this paper, an approximate non-FEA-based simulation method, based on a mesh morphing approach, has been the subject of a case study to evaluate its acceptance and applicability. It involved a focus group and individual interviews with engineers and project managers from two companies within the automotive industry. The study shows that providing the possibility to perform visualisation activities in the early phases is highly sought after, both on an engineering level and on a management level. Furthermore, a number of application scenarios for this type of approximate method were proposed. The study also identified strengths and risks of visualising the effects of geometrical variation in this way.

Non-FEA-Based Method as Means for Knowledge Based Assessment of Perceived Quality

Fit and finish of vehicle split-lines is one important contributor to the final Perceived Quality (PQ) of the product. To achieve high PQ of split-lines, effects of geometrical variation has to be considered. In early phases of the development chain the geometry models used for simulation and visualization have low level of detail. This limits the possibility to perform certain simulations that rely on a more complete detailed design. Consequently, alternative methods have to be considered to predict and simulate possible outcome in early phases concerning PQ issues. This paper proposes how an existing non-FEA-based deformation method can support virtual assessment of the PQ of split-lines in early phases. The method is based on mesh morphing and has been implemented in a CAT-tool (Computer-Aided Tolerancing). Its strength lies in the simplicity of generating deformed shapes. The paper specifically focuses on how available knowledge regarding issues from previous projects can be used as input to the method, to predict part deviation and part behavior. The paper further presents industrial examples where the method has been applied. The results show that the proposed technique can be used as a complement to other simulation tools in early phases, where low level of detail on geometries limits the possibility to perform FEA (Finite Element Analysis) based simulations.

Using Morphing Techniques in Early Variation Analysis

Today, in order to be competitive in a fierce global car market, higher demands are placed on the Perceived Quality (PQ) of the products. The end customers visual impression of fit and finish are one of several factors influencing the overall PQ. When assessing the PQ of split-lines, the assumed geometric variation of the ingoing parts is an important prerequisite for trustworthy visualization and for correct judgments. To facilitate early decision making in conceptual phases, new demands are set on virtual tools and methods to support the engineers. In this study, a method for early evaluation of the impact of geometrical variation on PQ of split-lines is proposed. Starting from an exterior styling model, mesh morphing techniques have been used to distort the exterior model according to measurement data acquired in running production. Morphing techniques have also been used to adopt previous structural design solutions onto the new styling, in order to make an early assumption of the assembly stiffness. The used method is described and adopted in an industrial case. The study shows that the presented technique can be used to create continuous and correlated datasets. Non-rigid part behavior can be included in early PQ evaluations, even if final CAD/FEA engineering design models do not yet exist.