Winter semester 22/23: The number of participants is limited to 24. Registration takes place via the ISIS course.
The current increasing complexity of global value networks and logistics activities as well as the new connected, intelligent and automated technologies increase the availability and use of data and the resulting strengthening of data-driven or evidence-driven management. This development is accompanied by increasingly powerful quantitative and data-based methods, which are mainly taken from statistics and machine learning. In addition, the use of these methods is constantly increasing in usability, which is due to the growth of data processing tools and their constant improvement. As a result, the use of statistics and machine learning in business management is always being promoted. The use of quantitative and data-based methods to find and solve problems in an economic context and for business tasks is called business analytics. Applied to logistics, it is referred to as supply chain analytics. The introduction of supply chain analytics has already made a significant contribution to improving business performance in some companies and further high expectations are being placed in it.
Business analytics is divided into three areas: (1) Descriptive (2) Predictive and (3) Prescriptive Analytics. Descriptive analytics goes beyond the aggregation and summarisation of data that is often mistakenly understood as such. The goal of descriptive analytics is to gain an understanding from data about fundamental phenomena, processes, relationships or causal causes that have led to past business events and outcomes. Predictive analytics aims to predict value that is not yet known. This can be future values that have not yet been realised or values that are not accessible to a company. A distinction can also be made between predicting continuous numerical values or categories. Finally, prescriptive analytics aims to provide recommendations for action in order to achieve a desired intention. All three areas should support decisions with their results.
The aim of the course is on the one hand the application-oriented teaching of methodological competence with regard to quantitative analyses using the example of problems from logistics. In the course, the methods are taught on the basis of problems with required decision support and the critical handling of the results and their uncertainty is focused on. The basics of quantitative methods enable the students to subsequently acquire further knowledge themselves and to be able to critically evaluate the use of the methods in everyday professional life. As an application-oriented course, the course is designed to enable students to apply the methods to problems. In-depth theoretical basics of statistics, optimisation or machine learning are not covered. On the other hand, the course is intended to impart management competence with regard to the use of quantitative methods. Students should understand when the use of which method is sensible and goal-oriented and recognise the limits of the methods. Within projects with the aim of decision support, the students understand the prerequisites for analytics, the necessary interlinking of methods to solve problems and tasks and the subsequent steps to bring about the desired decision support.
The contents of the course cover five topics:
The analysis methods are introduced and taught using Excel and then deepened using R programming for the analysis of larger data sets. An introduction to R programming is part of the course. The methods are presented using a gamification concept.
The examinations of the course are 3-4 homework assignments in groups of 2 (50 marks) and a case study in groups of 4 (50 marks).
The course requires a computer to work on the assignments with an installed version of MS Excel (free for TUB students) or LibreOffice (free of charge) as well as of R (the additional installation of R-Studio is strongly recommended) (free of charge). R is used with R Notebooks (equivalent to Jupyter Notebooks in Python) to enhance the learning experience. Where possible, students are encouraged to bring their own computers. The lecture room (H 9107) is limited to 12 computer workstations.
No explicit courses are required. However, prior attendance of Statistics I, Operations Research I, Introduction to Computer Science I and Fundamentals of Logistics or equivalent courses is recommended.
6 ECTS points are given for successful completion of the course.
The homework deals with methodical tasks to be solved in R and Excel as well as text and visualisation tasks. These are solved in groups of 2. The tasks are submitted in writing.
The case study comprises a comprehensive project that is to be solved in groups of 4 and then presented.