Quality Engineering (10 CFU)

Course objectives and description:

The course focuses on quantitative methods for modeling, monitoring and improving quality of products and services in actual productive scenarios, characterized by multi-dimensional quality indicators. Techniques for both continuous and ordinal quality data (e.g., survey data) are presented.

After successfully completing the course, students should be able to do the following:

1. Understand the philosophy and basic concepts of quality monitoring and improvement.

2. Extract relevant information from complex, high-dimensional quality data

3. Identify models to predict the expected pattern of quality indicators

4. Design and use appropriate tools to design and manage quality in industrial and service scenarios.

Course content:

1. Quality engineering for product and services: the Six-Sigma roadmap and related tools. Statistical Process Control (SPC) or Statistical Quality Monitoring: main assumptions and limits.

2. Quality modeling

• Standard assumptions (i.e., normality, independence) and related tests.
• Modeling patterns via linear regression
• Modeling autocorrelation via ARIMA models
• Modeling multivariate data and summarize related information via principal component analysis
• Modeling survey data: Categorical and ordinal data

3. Quality monitoring of continuous variables

• Traditional control charts for the mean and the variance
• Control charts for autocorrelated data: Problems of traditional control charts for autocorrelated data; Model based and model-free approaches for quality control of autocorrelated data.
• Design and managing multivariate control chart for the mean and the variance.

4. Quality monitoring of discrete, categorical and ordinal data

• Service Quality and customer satisfaction
• Binomial and multinomial control charts
• Control charts based on Logistic and ordinal regression

5. Toward zero-defect manufacturing: process quality and product specifications. Capability analysis. Univariate and multivariate control charts for small shifts (EWMA, CUSUM) .

6. Acceptance Sampling and Quality control – Lot-by-lot acceptance plans for univariate and multivariate data

7. Quality Improvement – The role of improvement in the six-sigma roadmap. Quality improvement via empirical model building (hints).

Course activity:

-          60% Lectures

-          40% recitations (esercitazioni) using realistic data set and specific sw (Minitab, Matlab)

-          Team Project (optional)– students will be required to develop their own strategy for quality modeling, monitoring or improvement dealing with reference to a real case study.

Textbooks:

• L. C. Alwan “Statistical Process Analysis” – Irwin Mc Graw Hill
• D.C. Montgomery: Introduction to Statistical Quality Control - Wiley
• E. del Castillo: “Statistical Process Adjustment for Quality Control” – Wiley
• A. Agresti: “Analysis of Ordinal Categorical Data” - Wiley Series in Probability and Statistics

Additional scientific material will be provided

Exam:

• Written exam using appropriate sw (minitab, matlab) - mark: 30/30
• Project discussion (only if written exam >18 - required only if the team project has been selected): maximum increase of the final grade 3/30
• Oral exam – optional (only if written exam >18): maximum increase of the final grade 3/30