Design of experiments and statistical quality controlLaajuus (3 ECTS)

Objective

The student understands in what kind of problems mathematical models help to solve the problem. The student knows how to choose appropriate model types for different problems. The student understands the importance of experimental design and the nature of experimental error and knows how it affects the interpretation of experimental results. The student knows that the choice of a statistical analysis depends on the chosen design. The student knows the most common graphical tools and statistical analyses (statistical tests, ANOVA, regression). The student knows when Excel is an adequate tool for analyses and when statistical software is required.
-----
The student is able design experiments in typical problems of R&D or quality control in bio- and food-engineering. The student is able to analyze experiments that have been conducted according to his/hers design. The student is able to report his/hers results using interpretations given by statistical analyses and graphical representations. The student is able to use Excel in making experimental plans and in simple statistical analyses of designed experiments.

Content

Empirical vs. theoretical (mechanistic) modeling. Variable types in mathematical models. Classification of mathematical models and their applicability in different problems of bio- and food-engineering. Basic statistical tools of modeling. Experimental errors, modeling errors and residuals. Designs for quality control problems. Factorial and central composite designs. The idea of experimental optimization. Graphical presentation of experimental data. Analysis of variance (ANOVA). Regression analysis. Response surface analysis.
-----
Fractional factorial designs, Taguchi, Plackett-Burman and related designs. Multi-response optimization.

Qualifications

Basic course in statistics

Further information

Class room teaching: 14 h
Computer labs: 21 h
Project: 15 h
Exam: 3h
Student individual workload (workload analysis not carried out): 27 h
Total: 80 h
Follow-up of the student workload analysis performed: -