In this paper, collars are classified into three basic types - stand-collar, roll-collar and open-collar according to their structure. A geometrical model is established to simulate these collars and methods of drafting patterns are considered too. The collar shape was modeled as three-dimensional Bezier patches. Changing the position of control points of these patches can interactively control the collar shape. Triangular surface developing method is employed to make the pattern for stand-collar. For roll-collar and open-collar, flat pattern method using the data calculated from three-dimensional coordinates of control points are employed. With these methods, an interactive collar designing system is developed to simulate collars and automatically make patterns. Considering practical usage, we reproduced three types of collars with the patterns drafted by the system, and observed these reproductions congruent with simulated ones well. It is demonstrated to be more reliable, efficient and accurate than the method of handwork.

View full abstract

In curve design, it is important to grasp and control the whole characteristic of shape that emerges from combinations of shape elements. However, it is difficult that only conventional shape-information expresses the whole characteristic of shape. Therefore, a new shape-information that could express the whole characteristic of shape and a design-support system using it as a design guideline are desired.
In past research, the authors have defined the statistical amount of shape (SAS) from information entropy and expectation value. Principal component analysis of simple curve profiles, such as geometric figures, was carried out using conventional shape information and SAS. Curvature entropy and curvature expectation value were included in the independent main component, and the relationship between SAS and various types of recognition information in the simple curve profile was analyzed. The correlation between curvature entropy and “complexity”recognition information was about 0.7. Curvature entropy is thus unlike conventional shape information. It can also express “complexity”characteristics in a simple curve profile.
In this study, we described shape-generation method using curvature entropy and genetic algorithm, and applied this method for the design of the automobile side view. As a result, it was confirmed that curvature entropy could express the whole characteristic of shape in a curve profile at the product level, and the possibility of the design-support system using curvature entropy was shown.

View full abstract

Reduct of the rough set theory is recently used as a mean of knowledge acquisition concerned to a cause-effect relationships consisting of attributes of an object as a cause and of classification as an effect. The data of the rough set theory is limited to categorical data. However, kansei data are often ambiguous. Accordingly, a certain observation value might not conform to a certain category but might only conform to within some grade. In order to make the knowledge acquisition be more agreeable to the actual condition, this paper proposes an algorithm of reduct calculation for the categories added with degrees of conformity which is appropriate to the expression of attributes related to kansei.

View full abstract

The function of artifacts depends on Ba. Ba is defined as elements of human, artifacts and surroundings and the relationships between each of these elements. The different relationship generates the differences in function of artifacts.
Generally, artifacts are used in diverse Ba. But many artifacts are designed under average Ba at present. Therefore, this diversity of Ba results in low evaluation of artifacts under particular human or surroundings.
In order to solve this problem, the objective of the present study is to construct a method for stable evaluation under diverse Ba. First, a method of robust design was proposed. Using objective function including average and variance, the proposed design method can adapt to diverse Ba and features modeling Ba and simulation using Ba model.
Secondly use of this method was confirmed by application for the railway vehicle swing-seat design. Swing-seat has a function, which prevents hip sliding by increasing cushion angle as seat back tilts. In the present study, the diversity of sitting postures and differences in physical constitutions were defined as noise factors. Optimizing design solution (relationships between back angle and cushion angle) for the object variables (hip-sliding force) was attempted. The simulation using Ba model, which was constructed as a dynamic model of human body and seat, was performed.
As a result, optimum relationships between back angle and cushion angle were obtained using the proposed method. It was proved that the proposed design method for diversity of Ba could be useful.

View full abstract

How the visual load and/or the conversation load influence functional visual field (FVF) during driving was studied experimentally. Red or pink circular target was briefly presented somewhere on a movie scene from driver's view in a car following situation, and the observer's task was to indicate the target position. The visual load for the observer was to read out the two alphabet letters presented in the center of the visual field. The conversation load was to answer to an easy question orally. The FVF narrowed with the visual load as well as with the conversation load, while it markedly shrinked when both visual and conversation loads were imposed simultaneously.

View full abstract

The use of quantitative design methods for KANSEI is an effective way of determining an accurate design solution in a short period of time. They are now being studied in conjunction with advances in CAE (Computer Aided Engineering), CE (Concurrent E ngineering), and other systems for application to product development.
This paper first introduces four examples of quantitative design methods for KANSEI from automobile research and development. The examples are “Deign Method for Seating Comfort”, “Coordinated Illumination System”, “Method for Evaluating Roominess in a Vehicle's Interior”and “Switch-Operation-Feeling Evaluation System”. It then exmines the features of these design methods and finally explains the need for systematic development and application in the form of a “gray box model”that combines quantitative design methods with qualitative design methods.
At present, there is a demand for placing high value on KANSEI in automobile development, and on the other hand, a great need for efficient product development from the point of view of corporate management. The appropriate selection of KANSEI design methods in terms of both results and efficiency is becoming a major issue in automobile development. To carry this out successfully, the application of KANSEI design methods must be strategically addressed with future plans made clear.

View full abstract

With the explosion in the amount of information on Internet, finding desired information put additional cognitive and repetitive burdens on the user. In order to overcome these drawbacks, we propose a layer of adaptive and collaborative agents between the layer of users and the layer of distributed information sources like Internet. The personalization is realized by learning the user model and using it at query formulation. Machine learning and information retrieval techniques were utilized to learn the user preferences and to provide support for well-formed personalized query reformulation. However, for learning agents working individually, they face two problems: (i) serendipity, i.e. they cannot deal properly with situations previously unseen in the past; and (ii) cold-start: they spend some time to relearn about new situations.
In order to deal with these problems, we add a layer of collaboration between the agents, where the selection of peers is based on the trust relationship among them. The collaborative aspect is obtained by exchange of information learned by the individual agents.

View full abstract

The science of identifying elements for corporate success through manipulation of accounting information is, in general, referred to as Business Analysis. The du Pont System is a representative method employed in business analysis in which the quality of management and managerial problems are identified by analyzing correlations among various accounts by means of the company's return on investment as the core parameter. While it is more or less long established practice for ROI to be the primary means of assessing corporate performance, today there can be a variety of methods for dissecting the parameter. In this study, the financial indexes for Japan's 1050 representative manufactures during the 15-years period from 1981 to 1995 were analyzed through the traditional method of the du Pont System, such statistical method as Correlation Analysis, and also through simulation byNeural Network (neural approach). It was deduced from the findings that added-value analysis indexes are more useful than the others.

View full abstract