In order to educate and train measurement techniques for students, the courses, including calibration of measuring instruments and evaluation of measuring error, were put as the core of the curriculums for the Measurement Engineering Department of our university. It was intended to let the students be exposed to the area of measurement techniques. In the process, the method of evaluating the measuring ability of students was discussed. Further study was made to predict the measuring ability of students. The students participated in the measurement and answered questionnaires so that the characteristics of students may be identified.

The new sheet-type cleaner provided by changing slurry into a sheet, is introduced to a market to meet the customer's requirement for good usability of a washing cleaner. This study applies a slurry cleaner of a dispersal system in which cleaner particles disperse in the liquid component including mainly liquid surface-active agents. The cleaner is characteristic of non-Newtonian fluidity,whereby the fluidity become unstable to lead to variation of the producing application process. The study's object is to design the composition of a slurry cleaner with smooth fluidity and make the process design even when concentrated cleaner particles are mixed. To achieve the object, the relation between shear rate and shear stress in the fluid state of the slurry cleaner is surveyed and the main functions of fluidity is studied on basis of a formula for the Casson. Since the shear history and the transient response to inputting steps may bring variations of the evaluation, the elapsed time recorded after the slurry is prepared was regarded as the noise factor. In the study implemented according to the orthogonal array L18, the preferable reproducibility was provided in the S/N and the sensibility of the checking test, and further the variation of fluidity in the optimal condition was reduced to approximately one-half. As a result to be expected, the application process becomes stable, and some functions are enhanced to the sheet-type cleaner provided by mixing concentrated cleaner particles and further the cost reduction is achieved.

Urinary incontinence is deeply related to the rehabilitation of brain disease patients, and it is an important matter of concern for clinical nursing. The approach of pattern recognition reported in this paper is to predict the urinary continence recovery four weeks after the surgical operation or stroke. The prediction is made based on the information including the attributes of the patients, type of brain disease, the level of consciousness (i.e., eye opening, verbal response, or motor response) and the level of continence one week after the operation or stroke. From the aspect of cost for prediction, the item selection method using orthogonal arrays was used. In the past study, items were selected in such a way to maximize the difference in Mahalanobis distance between the two groups of patients: the one, which recovered urinary continence after four weeks, and the one, which could not. Since urinary continence recovery was evaluated by an eight level score, the recovery level was omega transformed to be used as the signal, and Mahalanobis distance in logarithmic scale was used as the output. Based on such input and output, dynamic characteristic was used for item selection. This study shows how the signal factor was set and how the results were evaluated for item selection using dynamic characteristic in MT system.

Before painting a vehicle, there are normally required the eight processes; pre-rinse,pre-cleaning, cleaning, water washing, surface activation, conversion treatment, water washing, and purified water washing, where the high equipment material and management costs are needed. For achieving the big cost reduction as the primary reason, an approach in designing the optimal composition of a new material was made, which enabled integration of the four processes from cleaning to conversion treatment into a single process. In this approach to efficiency, the parameter design was employed on trial. As a result, there was fbunded the optimal composition of a material in a short period and at low cost, which ensured the robust processing capacity for the liquid agitation condition or the oil content quantity that may become a deteriorative factor in processing capacity, Here, the optimal composition was designed with the velocity ratio method as the parameter design, and the test conditions were determined with use of an L18 orthogonal array, to the inside of which the control factors including the main ingredients of the material and the processing conditions were set, and to the outside of which the noise factor to be defined by mixture of the liquid agitation condition or the oil content quantity of liquid were set, and further where the signal factor to be given by response time was set, resulting in that the test would be efficiently conducted.

Partial felting processing is one of the processing methods, in which the given patterns are printed on wool cloth by felt-resist paste the main constituent of which is thermo-sensitive polymers before conducting the partial felting process. As a result, the not-printed area becomes felt and the printed area does not become felt so that the material including both felt parts and not-felt parts is produced to provide variations. The partial felting processing has a problem in that the thermo-sensitive polymer to be used is expensive, In the last report, optimization of the felt-resist paste application was performed so that the paste cost was reduced to about one-third compared to the previous method. However, the further cost reduction is necessary for the practical use. Here, optimization of the whole processing method was studied as well as that of the felt-resist paste application, thereby reducing the cost to further approximately 40% compared to the result given by just the optimization of the felt-resist paste application.

As one of the approaches to help the students understand the contents of quality engineering, also to improve the ability of report presentation, an experiment on a turning process was conducted. The time spent for this project was 6 months, 5 frames a week. In the tuning process, two approaches were studied for evaluation; by transformability and by processing power consumption. It would be more preferable to evaluate the function by power consumption, but the studying to optimize, the processing condition was selected, since it was easier for the students to learn quality engineering for the first time in a short period of time. Based on the results of this experiment, the points that should be focused in the learning process for a better understanding to the students were reviewed.

Efficient water feeders are a critical element in steam and hot water boiler systems. Overfeeding will cause the system to flood; not enough water will cause boiler damage due to overheating. Likewise, any other interruptions in a smooth feeding operation are not tolerated by the end user. As the dominant market leader in the manufacture of boiler controls, McDonnell & Miller's (M&M) mechanical water feeders are an integral part of heating systems world wide.

Field reports of "chattering" energized M&M to apply the Taguchi Robust Design technique in order to enhance the design and make it most robust against the field noise conditions. Chattering refers to a condition that manifests itself when the feeder valve is slightly open or slightly closed. A system harmonic is created, resulting in a loud, rolling noise through the heating pipes. This noise is considered by customers to be a significant nuisance and must be minimized or eliminated.

A dynamic Taguchi L18 experiment was performed. The experiment analyzed the level effects of six (6) control factors and three (3) noise factors. The "Zero Point Proportional" equation was used for the analysis. The inlet pressure to the feeder valve and cross-sectional flow area are utilized as dynamic signals. The flow rate through the valve is used as the response. The SNR (Signal-to-Noise ratio) of the present design is -8.87 dB and the predicted SNR for the optimum design is -2.97 dB. The SNR for the confirmation run is -4.25 dB. Experiment results showed a SNR ratio gain of 4.63 decibels compared to our present design. This indicates that a significant improvement to the chattering problem could be potentially achieved.