URL study guide
https://tue.osiris-student.nl/onderwijscatalogus/extern/cursus?cursuscode=2DD80&collegejaar=2025&taal=enDescription
As many business processes can be modeled using random variables, statistics is an important toolbox for Industrial Engineers to analyze a problem or situation using quantitative data. This course covers the basics of probability theory and statistics for Industrial Engineering.The course consists of 4 building blocks:
Probability Theory: Many processes or variables in IE can be described using either discrete or continuous distributions. This part covers how to deal with univariate as well as multivariate continuous random variables. It introduces concepts such as the density function, mean, variance, correlations and conditional probability distributions.
Point and Interval Estimation: This part deals with the first steps of using data and convers topics such as descriptive statistics, point estimation of mean and variance, as well as interval estimation of mean and variance, and the importance of the Central Limit Theorem.
Inference: This part of the course moves on to hypothesis tests, e.g. paired and unpaired t-tests, p-values, Type I and Type II errors, statistical power and size.
Regression: The third part of the course deals with linear regression models, both the simple and the multiple linear regression model are studied. This includes the Gauss-Markov assumptions, the concepts of bias and consistency, as well as the use of regression models, including standard linear regression as well as machine learning approaches.
Students will learn to recognize when to use which statistical methods and apply these in the context of IE problem solving. The course will not only cover the statistical theory, but will pay special attention to the application of the methods, the interpretation of statistical results and the translation to decision making in Industrial Engineering problems, e.g. in the context of statistical quality control in manufacturing or market testing of product innovations. The assignments cover both statistical simulation to assess the properties of estimators and application of statistical methods to real cases using Python.
Contents topics:
Continuous distributions. The concept of expected value and variance. Normal, exponential, Erlang-distribution. Weibull, lognormal and Gamma-distribution. Poisson-process. Normal approximation of Binomial and Poisson-distribution. Central Limit Theorem. Discrete simultanuous distributions, covariance, correlation. Independence of random variables. Linear combinations of random, especially of normally distributed variables. Estimation theory (unbiased, mean squared error. Confidence intervals (for mean, variance of a normal distribution and of a fraction), prediction intervals. Hypothesis testing (type I and type II error, p-value, choice of sample size). Two samples: confidence intervals for the difference of the expected values for paired and for independent samples. Simple regression and correlation. Multiple regression: estimation of parameters, confidence intervals and prediction intervals; model adequacy checking; model building.
Objectives
The aim of ‘Statistics for IE’ is to provide industrial engineering students the knowledge and practical skills to address business and research problems using quantitative data and statistical analyses.More specifically, students will:
- Understand the concept of continuous probability distributions including expected value and variance. They will know the important continuous distributions and will be able to apply the correct distribution in a given situation.
- Understand the use of two-dimensional stochastic variables in the case of simple discrete distributions.
- Understand the principles of statistics, pply estimation theory in various problems, andunderstand the concept of confidence intervals and construct them in various situations.
- Understand the concept of hypothesis testing and apply hypothesis testing for making decisions.
- Use the chi-square test to check distributional assumptions.
- Test hypotheses and construct confidence intervals on the difference of means of two normal distributions for paired and independent samples.
- Use multiple regression techniques to build empirical models, and assess regression model adequacy.
- Understand differences in linear regression approaches and machine learning approaches to model data.