Design of Experiments – DoE

//Design of Experiments – DoE
Design of Experiments – DoE2019-09-02T16:57:57+00:00

Training Information

DoE – experimental design DoE (Design of Experiments) is a structured, statistical planning and execution of processes to obtain relevant product and process parameters that influence their respective influence on output variables of interest. DoE provides a data model on the cause-and-effect structure of products or processes.

From this, characteristics and interactions as well as the optimal settings can be determined. Get an overview of DoE’s practices and techniques in a mix of training and workshop exercises. The learned knowledge can be profitably converted into operational practice.

Duration3 days from 9am to 5pm
Date
Venue
Cost excl. VAT
QualificationAfter completing the seminar, each participant will receive a certificate of attendance.
Minimum participants
PrerequisitesEach participant must have a notebook with the software Microsoft® Excel from 2003, MINITAB® from R14 and Adobe® Reader® from 7.0. The software MINITAB® can be obtained from us.
Print Description
Print Friendly, PDF & Email

Objective:

With the method of statistical experimental design, you learn to gain maximum results about the effects of influencing parameters and with as little effort as possible. It will show you solutions how to improve the quality of your products and processes efficiently.

Target group:

Employees and executives from the fields of engineering, quality management, production, CIP, technicians as well as R & D from manufacturing companies and service companies.

Seminar content:

  • Introduction to DoE – Design of Experiments – Overview
  • Basics of statistics and regression and analysis of variance
  • Elementary capability metrics
  • Calculation of required sample sizes and experiments
  • Preparation of experimental plans
  • Systematic derivation of the relevant input information for DoE’s
  • Review of the understanding of Y and X variables
  • Dealing with noise variables
  • Handling of continuous and discrete variables
  • Setting up a roadmap for planning a DoE
  • Selection of factors
  • Determination of the factor levels, determination of the test room
  • Selection of the design on the basis of existing (data-based) knowledge
  • Determination of the number of replications, sampling theory to DoE
  • Preparation of documentation of results and accompanying information
  • Development of suitable templates
  • Consideration of the experimental environment
  • Coordination with stakeholders, e.g. in the production
  • Influence of measuring systems on a DoE
  • Experiment designs and evaluation
  • One factor method (OFAT = One Factor at Time)
  • Full factorial design plans (2 ^ k Full Factorial)
  • Setting up a DoE test plan
  • Effects of major factors and interactions
  • Graphical representation of the results (main effect diagram, interaction diagram, cube diagram)
  • Calculation of the effects
  • Determination of statistically significant effects
  • DoE – exercise on an example
  • Partial factorial designs (fractional factorial designs)
  • Screening designs to determine the key factors
  • Statistical analysis of test results, regression modelling
  • Residual analysis with interpretation
  • Further use of a model (output forecasts, “best settings” in production
  • Desirability – Functions
  • Recognizing and interpreting critical states in (product and process) designs
  • Exploring the performance limit of a design
  • Handling, weighting and meaning of several Y-variables
  • Simulation options, Response Optimization
  • DoE and variation reduction
  • Response surface designs
  • Approach and theory of the central composite design (CCD) including variants
  • Box Behnken Design
  • D-optimal designs (approach, benefits and risks)
  • Orthogonality and rotation
  • Theory for the statistical evaluation of RSDs
  • Handling a complete roadmap for analysis and modelling
  • Non-linear desirability functions
  • Evolutionary Optimization (EVOP)
  • Outlook
  • Mixing experiments
  • Application situations and approach
  • Variants
  • Statistical modelling
  • Taguchi Designs
  • Application situations and approach
  • Planning, selection and integration of noise factors
  • Statistical analysis
  • Analysis and interpretation of the robustness of designs
  • Build a comprehensive DoE roadmap with all branches
  • Systematic expansion of acquired DoE knowledge in the company
  • Setting up and targeted use of process algorithms (transfer functions)
  • Optimization of process chains, production lines etc.
  • DoE in DMAIC and DFSS projects
  • Targeted use of DoE to set tolerances
  • DoE to Data Mining interface
  • “DoE” in interaction with “neural networks” or decision trees
  • Tips and exchange of experience

Scope of services:

  • Consistent case study
  • Comprehensive training documents (in paper form and as .pdf files)
  • Numerous data files, tools for practical exercises
  • Room for exchange of experience and expert knowledge transfer