One of the most important problems in constructing fully automated machining systems is the need to reduce the interruption of the cut due to the sudden tool fracture during the cut. High reliability is prerequisite for the designing of efficient and economical automated machining systems. This paper presents a statistical model to analyze the reliability of automated machining systems. The statistical model based on tool life distribution caused by wear or fracture is developed. Critical fracture rate is newly defined to reduce the interruption of the cut due to the sudden tool fracture.
In this paper, we propose a reactive Kanban system for the unstable changes of product demand. In the proposed system, the exponentially weighted moving average chart is utilized for detecting the unstable changes of product demand, and the appropriate number of Kanbans is calculated on the basis of the detected changes. By adjusting the number of Kanbans dynamically to the changes, the proposed system is able to respond to the changes. The performance of the proposed system is analyzed by simulation experiments, and it shows the effectiveness of the proposed system.
Industrial Teamwork Dynamics (ITD) has been researching from positivism point of view through Social Survey in Industries. On the secondary research, this paper applies and verifies ITD as new paradigm using Group Dynamics of Social Psychology for group behavior at bottom line workers into the Japanese ladies innerwear sewing maker. The work types as work team were divided into three categories, group type as flow-line (sewing by machine), individual and crew type (checking textile and cutting), individual only type (inspection). The teamwork level was defined and extracted by PCA. The team productivity was confirmed as work efficiency, achieved ratio of actual work time to standard work time in each work team. ITD certifies group type refined better teamwork level than others according to G-P analysis (best ratio groups : 151 persons, bottom ratio groups : 173 persons including three work types) by ordered work efficiency ratio on every work type.
Kansei engineering is a technology for translating feelings into product design. There are several methods for evaluating psychological responses of human beings, such as the semantic differential(SD) method and pairwise comparison methods. In this paper, a projection pursuit regression method for paired comparison data is proposed, which is based on the membership function of fuzzy set and eigenvector of covariance matrix. The dimensionality of predictors is reduced to one in order for good use of human ability of instantaneous pattern discovery.
It is important to study on the usability for the vending machine in order that we think about man-machine interfaces for aged people, because the vending machine is the machine which many aged people use every day. So from the study like this we can accumulate the basic data which can be applied to various machinery in the future. Improvement on the directions for use in the experiment shows both the decrease of purchase time and that of hesitant behavior. This means an improvement on indications like this, in other word, an improvement on the software part of indications, helps the users who utilize vending machines. In the actual situation, there is no directions of explaining how to use the ticket vendor machine. A communication between the machine and users is not enough now. We hope that those who are engaged in designing the vending machine will take into consideration in the future the improvement on the software part of indications for users.
Recently information technology is being rapidly introduced into office work environments. At the same time, percentage of older people in the population is growing and will affect the structure of labor force. In this vein, training of middle-aged and elderly workers in computer operating skills is becoming more important. One of the training methods that is expected to be effective for such technology training is exploratory training. An experiment was carried out to investigate the validity of exploratory training in comparison with traditional instructional training when applied to the training of middle-aged and elderly persons in computer operating skill. Both young and older subjects participated. Subjects attended either exploratory or instructional training courses to learn skills to perform information search and input tasks in personal computers. Then the subjects performed some sample tasks and performance indices such as operation time, task accuracy, task cognition, and exploratory behavior were measured. The results show that validity of exploratory training for older subjects is different between people who are currently employed and those who are not. From the view point of operation time and task cognition, exploratory training seems to be more effective for people who are currently employed, while instructional training seems to be effective for people who are not. From the view point of task accuracy, instructional training is superior for both younger and older, currently employed and not employed. Training validity and task accuracy are not significantly different between age groups, while operation times differ significantly. The results suggest that effectiveness of exploratory training is influenced by the trainee's knowledge even if it is not directly related to the task. Subsequently, the trainee's occupation and background knowledge need to be considered in relation with the skill to be acquired when a training method is selected.
This paper deals with a classification method for manual alphabet of Japanese sign language by image processing technology. An alphabet expressed by using the right hand as the finger letters is characterized from the condition of the fingers and the movement of a hand, Therefore the finger letters classify the finger condition and the hand motion into three classes for analysing an alphabet by image processing. The feature values of an alphabet is extracted as follows. First, the hand contour line extract from the time series hand images by two video cameras which were set up on side and on top. Second, the feature values of an alphabet calculate from the hand contour line. The feature values are the center and the width of wrist, the center of gravity of the hand, and the angle and the length from the straight line segment that connect the center of wrist and any point on contour line. An alphabet is judged by these feature values. Lastly, authors describe the classification method to apply to some images, and show some examples to demonstrate this method.