Hironobu Kawamura

Process Adjustment Control Chart for Simultaneous Monitoringof Process Capability and State of Statistical Control

In the production of chemicals, a process adjustment such as feedback control is frequently used to reduce process variability. It is very important to judge whether or not the adjustment should be done automatically because an automatic process control (APC) system requires a large capital investment. This paper presents the determination of the adjustment timing on the basis of the process capability, and control charts combining information about the state of statistical control and process capability are also presented for the judgment of adjustment timing. Practitioners can assess both the adjustment interval and the number of adjustments by simulation or trial using the presented method. Moreover, the information is very useful for judging whether or not the automatic adjustment system should be introduced.

Control characteristics: a case study on semiconductor manufacturing

Purpose – This paper aims to clarify adequate control characteristics for using a control chart on the basis of a case study of the low‐pressure chemical vapor deposition (LPCVD) process, which is one of the semiconductor manufacturing processes.Design/methodology/approach – The paper opted for a simulation study using the data generated by EWMA model and the real data obtained from the LPCVD process.Findings – The paper provides adequate control characteristics for control charts. It suggests that it is desirable to employ both the quality characteristic and the process rate for monitoring when the process was modeled by the EWMA model. Furthermore, if only one control characteristic is employed, then the process rate is the most adequate characteristic.Originality/value – This paper newly proposes the process rate as a control characteristic for control charts.

Proposal of Advanced Taguchi’s Linear Graphs for Split-Plot Experiments

Taguchi’s orthogonal arrays and linear graphs are convenient tools for the design of fractional factorial experiments, especially for practitioners. Taguchi also proposed how to use them in split-plot designs and prepared linear graphs for split-plot designs. For the orthogonal array of order 16, Taguchi proposed one which is called L16 orthogonal array. Taguchi presented 18 linear graphs when a L16 orthogonal array is used in split-plot designs. Those linear graphs are capable of showing main effects of whole plots, subplots, sub-subplots, and so on, but they are not capable of showing interaction effects of plots of different levels. Also, those linear graphs do not cover all the possible designs, and there exist a lot of other linear graphs that can be applied when using L16 orthogonal arrays. The primary objective of this paper is to propose an improved version of linear graphs. Another purpose of this paper is to investigate how to list all the possible linear graphs that can be applied when using L16 orthogonal arrays. A proposal is made and many new linear graphs are presented.

Capability of Detection for Poisson Distributed Measurements by Normal Approximations

In the analysis of very small components, it is very important to know what concentration or amount of the analyte can be detected by the measurement method. When we determine whether the analysis sample is the same as the basic state, the capability of detection for the measurement method is defined as the amount that can be detected. The ISO 11843 series standardizes the capability of detection. ISO 11843 series has not described the capability of detection for Poisson distributed measures. The measurement results occasionally are count data. When the capability of detection can be exactly measured, it can be derived numerically, but it is very difficult to be derived analytically. There are various approximation methods having their own distinct features for a Poisson distribution. However, so far it is not known which approximation is best from the viewpoint of estimating the capability of detection. In this paper, the evaluation of the capability of detection for Poisson distributed measurements is described. A number of approximation methods are proposed from the viewpoint of the capability of detection for the measurement method. The best approximation method is then compared with the exact method.

Factor assignment based on linear-by-linear interaction for a three-level supersaturated design

A supersaturated design can be used to screen many potentially relevant factors with a smaller number of runs; thus, the resulting matrix has more columns than rows. A three-level supersaturated design is useful when second-order effects also need to be determined. However, experimenters using three-level supersaturated design may actually miss the main effects of factors because the columns of the matrix are not all orthogonal to one another. This paper shows how to assign factors to three-level supersaturated design that considers low-order interactions. Specifically, factor assignment that avoids confounding all main effects with all l × l interactions is presented for 27 × 61 and 24 × 28 supersaturated design matrices. The usefulness of the assignment columns created using this method is shown by comparing the presented factor assignment method with the existing one in terms of level of non-orthogonality.