|Teaching method||Contact hours|
|Course coordinator(s)||dr. R Haijema|
|Lecturer(s)||dr. R Haijema|
|dr. J Valente|
|Examiner(s)||dr. J Valente|
|dr. R Haijema|
Language of instruction:
Assumed knowledge on:
Prior to the course students should be familiar with Programming in Python (e.g. by completing the course INF-22306).
Some prior experience in supply chain modelling, simulation and/or programming is beneficial. Students should have a basic understanding of sampling from a probability distributions.
INF-31306 Engineering and Management of Information Systems
INF-33806 Big Data
ORL-30806 Operations Research and Logistics
ORL-30306 Decision Science 2
These days research and education in the field of Operations Research and Logistics are shifting, spurred by the growing availability of data in supply chains and recent advances in machine learning and optimization methodologies. Modern supply chains get more and more digitalized, and more and more data is recorded and stored. Data analytics sheds light on the value of such data in answering business questions. Examples are:
- can one predict failures of products or machines?
- can one predict the return of products by consumers?
- what data is needed to better predict the demand?
- by how much can one reduce food waste by better predicting the demand or a product’s shelf life (quality)?
Operation researchers usually start with analyzing the problem to develop a parameterized model which need to be filled with data. The decision made are based on a model that is set by logical and causal relations between variables and may include uncertainty through stochastic variables. Recent machine learning approaches start with data and learn from the data a model that fits (statistically) best. The advantage of such models is they may discover patterns not detected before, and thus may support better decision making. The disadvantage of such models is that the (statistically learned) model may not describe well the causal relations found in practice, and is thus hard to understand, and risky to apply to new data. Supply chain managers should thus be aware of the possibilities and limitations/risks of machine learning methods.
In this course we give students an overview of the most relevant machine learning techniques. We focus on descriptive and predictive analytics using methods like k-nearest neighbour, linear regression, ridge, lasso, logistic regression, support vector machines, and the basis of neural networks, and gradient boosting for so-called classification and regression. We will not go into the mathematics of the algorithms of machine learning. Instead we focus on introducing the methods and the good use of such methods and potential application is a supply chain management context. Students gain hands-on experience with simulation and machine learning tool boxes available in Python.
After successful completion of this course students are expected to be able to:
- explain the role of data in (food) supply chain management;
- apply data science techniques (classification, regression, forecasting) to a provided data set;
- create a (computer) simulation model of a part of a (food) supply chain;
- quantitatively assess the impact of applying data and data science technique to a particular decision problem within a supply chain.
- self-study: to prepare for and to complete practical sessions;
- active participation during computer practicals to acquire both the skills in Python to gain a better understanding of the theory in the book;
- short lectures provide the students with a review and preview of the methods and applications that are studied during practicals;
- in class tests and assignments: to learn about SCM application and to test students individual skills and knowledge.
The examination will be based on in class tests and assignments during the course, and a short final exam.
- Assignment 1 (20-25%)
- SCM practicals/presentations (10-20%)
- Assignment 2 (30-35%)
- Knowledge/Theory test (20-30%) = short final exam
Active participation to the practicals is required to pass for the course. Exact weights and minimum grades can be found in the Course Guide.
- Müller, Andreas C., and Sarah Guido. Introduction to machine learning with Python: a guide for data scientists. O'Reilly Media, Inc., 2016. EAN 9781449369415.
- practical exrecises with instructions (on Brightspace and via Google CoLab notebooks)
- links to website, and
- references to literature (on Brightspace).
|Restricted Optional for:||MME||Management, Economics and Consumer Studies||MSc||A: Spec. A - Business Studies||2AF|