NOSKIEVICOVA, Darja
Dept. of Quality Management, Faculty of Metallurgy and Material Engineering, VSB-Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic, darja.noskievicova@vsb.cz
Abstract: in recent years there has been a strong movement for greater emphasis on TQM which statistical thinking has been a major contributor for management development. Continuous process improvement, people involvement in it and prevention form fundamental principles of the TQM philosophy. Their realization needs permanent evaluation of the process behavior using data collection and data processing. It can't be done without the application of statistical methods. The acceptance and absorbing of statistical thinking is the main pre-condition of successful application of the statistical methods.
The acceptance of phenomena "process variation" is the basis of statistical thinking. We must consider variation to be immanent process property that is needed to be permanently quantified, analyzed, controlled and reduced through never-ending finding out of variation causes and accepting suitable effective corrective actions. It provides opportunity for never-ending process improvement.
The acceptance of statistical thinking doesn't mean only willingness to learn methods algorithm and effort to implement them into practice without deeper analysis of pre-conditions for effective application of selected methods. Unfortunately such limited statistical thinking is not unusual in the practical applications of the statistical methods in our companies. In many cases it results in lower efficiency of the applied methods or even in their failing and loosing of the employees trust in the abilities of the applied statistics.
Complex access to statistical thinking requires in addition verification of meeting of pre-conditions mentioned above, ability to select suitable method and to interpret obtained results correctly. Very important aspect of statistical thinking is learning not only from bad situations but from good situations too. It is a large resource for the process improvement.
This paper deals with the analysis of study programs and curricula for every form of studies guaranteed by the Department of Quality Management in point of the education for statistical thinking. The special stress is put on the subject "Special Statistical Methods".
Keywords: statistical thinking, statistical methods, processes, variation, process improvement
Statistical thinking must not be reduced only to specialists in statistics. Everyone who wants to realize never-ending process improvement must be able to use it and must use it. Statistical thinking is a philosophy of learning and actions based on following fundamental principles:
The acceptance of phenomena "process variation" is the basis of statistical thinking. If we want to think "statistically" for the first we must accept the life true that variation is the immanent property of every process. It means lack of repeatibility of the process (for instance every design engineer knows that he is not able to obtain all rotor shafts identical as to their diameter). But the acceptance of variation phenomena is not enough from the point of view of the process improvement. We must analyze variation causes to be able reduce their influence on the process and then apply correct actions leading to reducing variation of process with the aim of more consistent process outputs which require less inspection and testing. It can bring lower costs.
Process analysis is not possible without setting the main process and its outputs parameters their measuring, analysis and interpretation. This asks for the effective data collection and application of suitable statistical methods. But we must put the stress on the following facts as to the classical statistical access:
This passive approach does not directly lead to the process improvement.
The law of large numbers says that means are more tightly grouped than individuals. This is true as for human behavior as it is for the data: there is less variability in collective (team) thinking than there is among the ideas of individuals working separately. It means that
group thinking is usually better, less variable and more precise than individual thinking.
It needs effective leadership to elicit group thinking and keep its on target and accept the creative thinking of outliers.
From all ideas mentioned above we can derive following facts:
The Department of Quality Management was established in 1992 at the Faculty of Metallurgy and Material Engineering for the engineering branch of study "Quality Management". Since 1996 this department has been representing VSB - Technical University Ostrava in the European Foundation for Quality Management (EFQM) in Brussels. The department ensure:
The study programs fully meet the requirements of the European Organization for Quality regarding to EOQ - Quality Manager. The graduated should among others accept the idea of the process variability and its capability. It means that the education for TQM involves also clarifying the basis of statistical thinking and its fundamental principles, application of methods for identification of the problems and their causes, simple and complex statistical methods.
The following table contains subjects which wholly or partially contribute to forming of statistical thinking in the minds of the students.
|
Subject |
Parts of statistical thinking covered by subject |
|
Leadership and Communication |
team work methods principles of effective leadership |
|
Theory of Probability |
basic theory for understanding a analysis of process variation |
|
Mathematical Statistics |
Introduction to Statistical Inference |
|
Theory of Econometrics |
Special statistical methods (linear regressing models, Time series analysis) |
|
Design of Experiments |
Complex statistical tool for reducing variation |
|
Quality Planning |
tools for decision-making process (analysis of the causes of variation), for instance affinity diagram, interrelationship diagram, tree diagram, matrix diagram, QFD, FMEA) |
|
Computer Aided Quality |
software for quality data analysis and its interpretation |
|
Metrology and Testing |
Variation in measuring system Measuring of quality parameters Calibration |
|
Testing and Certification |
validation of testing process uncertainity of measurement in calibrations |
|
Special Statistical Methods |
methods for describing structure of processes understanding of variability (common and assignable causes) analysis of pre-conditions for effective application of selected statistical methods correct data collection principles, application and interpretation of the results of the methods for:
|
Table 1. Subjects and their contribution to the statistical thinking
In the frame of subject "Special statistical methods" the stress is put on the following aspects.
Understanding and description of the process
At the beginning of the course a big stress is put on the clarifying of the process, its structure and methods for description of its structure (flow charts).
Understanding of the basis of process variation
Really big stress is put on the understanding of the process variability - its substance, the analysis of it through differentiation between common and assignable variation causes, need for measurement of the parameters describing behavior of the process is discussed too.
Data collection
Effective ways for data collection (check sheets), methods for creating of random samples ( rational subgrouping) are discussed and trained.
Analysis and verification of the statistical pre-conditions for successful application of selected statistical methods
The acceptance of statistical thinking doesn't mean only willingness to learn methods algorithm and effort to implement them without deeper analysis of pre-condition for effective application of the methods. In the frame of the course "Special Statistical Methods" is put a big stress on the statistical pre-conditions of the suitable application for instance for
Analysis and verification of non-statistical pre-conditions
Not only statistical pre-conditions must be considered and assured for the effective application of the selected statistical and graphical methods. For instance in the frame of study of
must be considered.
Application of selected statistical and graphical methods
Many examples of the application of the selected statistical and graphical methods are trained. The learned methods are:
Check sheets, histograms, Ishikawa diagrams, Pareto diagrams, flow charts, scatter plots, control charts, acceptance sampling.
Interpretation of the trained statistical and graphical methods
A large emphasis is put on the final phase of the tool application - the interpretation of the obtained results. For instance:
Histogram is shown to be suitable tool for:
Scatter diagram is shown to be effective tool for the graphical representation of the existence or non-existence of two variables correlation.
Pareto diagram is shown as effective tool for the setting "vital few" factors from several points of view (frequency of factors, costs of factors, safety point of view).
Interpreting of the control charts from the point of view of the assignable causes effects is trained (tests of nonrandom patterns, re-calculated control limits, changes in a and b risks).
Statistical thinking is the gate to reduction of variability and never-ending process improvement. That's why everybody interested in the TQM application must understand it and apply it during his problem-solving process. The ways of incorporation of statistical thinking into the education process at the Department of Quality Management at the Faculty of Metallurgy and Material Engineering, VSB-Technical University Ostrava were discussed in this paper.
BRITZ,G., EMERTLING, D., HARE, L., HOERT, R. & SHADE, J. How top Teach Others to Apply Statistical Thinking. Quality Progress, June, 1997, pp. 67-79.
HARE, L.B., HOERT, R.W., HROML, J.D. & SNEE, R.D. The Role of Statistical Thinking in Management. Quality Progress, February, 1995, pp. 53- 60.
PETRIKOVA, R., NENADAL, R., TOSENOVSKY, J., PLURA, J. & NOSKIEVICOVA, D. The Education for Global Quality Management at Czech Universities and Companies. In 4th World Congress on Professional Development for Global Engineering Practice. Sydney: The Institution of engineers, 1997, pp. 121-126.