DIP60 Advanced Tools and Methods for Quality Control

Module code: DIP60


ECTS Credit Points:
20

Module Type: Compulsory


Year:
2nd

Language: Greek

Module general description:
Statistical Process Control
Introduction to statistical process control, chance and assignable causes, assignable causes and statistical process control, Process capability indices, accuracy and precision, use of time series diagrams in statistical process control, the use of hypothesis testing in assignable causes search, type I and II error, operating characteristic curve, in-control and out of control process, searching for mean shifts, time-series diagrams and histograms. X-bar carts, R-charts and s-charts. improving the sensitivity of Shewhart control charts, rational subgroup selection, applying the central limit theorem in control charts. Moving range and moving average, I-MR control charts, moving average control charts, EWMA and CUSUM charts, control charts for medians. Control charts for attributes, p-, c- and u-charts, sample size determination for p-charts, Demerit and Q charts, statistical process control for short production runs. Defects per million objects (DPMO) and first trial yield, 6 sigma program, quality control techniques for high yield processes, rejection control charts, acceptance process level, pre-control charts, ISO-9001 standard and quality control.
Use of the Minitab statistical package for case studies.


Design and Analysis of Experiments
Use of designed experiments for process control improvement, principles of experimental design (response variable, factor, level, treatment, experimental unit, background variable, blocking, noise factor, experimental error), completely randomized experiments with one factor, graphical analysis, one way analysis of variance, multiple comparisons, block designs and Latin squares, random effects, estimation of missing values, factorial designs with more than one factors, main effects and interactions, random effects models and mixed models. 2f factorial designs, 22 and 23 factorials, factorial designs and regression models. Blocking and confounding, fractional designs, resolution of a fractional design. 3f factorial designs, linear and quadratic terms of main effects. Use of the Minitab statistical package for case studies.

Procedures and Techniques for Continuous Improvement of Quality
Off-line quality improvement and control techniques, the usefulness of off-line quality control methods, the Taguchi approach for experimental design and off-line quality control, Orthogonal factorial arrays (designs), interaction tables, factorial designs for off-line experiments. Performance measures, the Taguchi approach for selecting performance measures, variability-control factors, target-control factors, data transformation in the performance measures approach, Box-Cox transformation. Taguchi’s orthogonal arrays, the multi-level formation, dummy-level, and virtual level techniques, the technique of composition and confounding, selecting optimal factor levels. Tolerance design, Taguchi’s approach  to tolerance design and analysis. Response surface methods and designs, special response surface designs, optimal experimental designs, the evolutionary operation method. Continuous improvement techniques, risk analysis, the Deming circle. Use of the Minitab statistical package for case studies.

Learning Outcomes: After completing this module, students are expected to be able to,
1. discriminate between random (common) causes and assignable (special) causes which are present in manufacturing process.
2. understand how Statistical Process Control (SPC) can be used to monitor a manufacturing  process and improve its performance.
3. apply quality control tools (control charts for variables and attributes, estimation of process capability indices) in a manufacturing process so as to achieve an optimum process level in terms of target value and variability.
4. understand and exploit a six-sigma improvement model.
5. indentify single-factor and multi-factor experiments as well as complete block experiments.
6. use experimental design theory (one or two way ANOVA, Latin squares etc ) to investigate the effects of one or more factors over a response variable (characteristic of interest).
7. discriminate between random and fixed effects models. Explain when each of them is applicable in a real data problem.
8. manipulate full and fractional factorial designs to perform statistical analyses of experiments involving factors with two or three levels.
9. employ Taguchi’s loss function approach to establish a value base for the development of quality products.
10. use Taguchi’s philosophy for off-line quality control by setting up appropriate inner and outer array designs and by analysing relevant performance measures, in order to determine the appropriate production process conditions so that the process and/or the final product is made robust against the effects of uncontrollable (noise) factors
11. exploit Taguchi’s lists of orthogonal arrays for setting up economical designs for experiments requiring the study of many multi-level factors.
12.  understand and apply response surface methodology for modelling the shape of a process’ or a product’s response.
13.  apply the methods of Evolutionary Operation (EVOP).

Subjects covered:
1. Statistical Process Control
2. Design and Analysis of Experiments
3. Procedures and Techniques for Continuous Improvement of Quality

Learning Material: The HOU publications can be viewed
here.

Teaching Method: Distance education with five Contact Sessions held at weekends during the academic year

Evaluation: Completion of five written assignments during the academic year, the average grade of which constitute a 30 percent of each student's grade, if a pass is obtained in the final or repetitive examination. Final exam grades constitute a 70 percent of the students' final course grade. More information is available
here.