Inhalt

Abstract:

This course provides students with a very practical, hands-on introduction to Data-driven Supply Chain Management (DSCM) using Machine-Learning (ML) techniques. Based on a specific example and dataset from practice, students will learn how simple and more advanced ML techniques (e.g. Neural Networks, Random Forests) can support decision makers in using extensive data to come up with better decisions in Supply Chain Management.

The course is structured around a single case example and a single set of data and will gradually introduce participants to fundamental and more advanced concepts of DSCM. In particular, students will learn how to build, employ, and evaluate simple and more advanced ML-models that can be directly used in practice. The individual lectures will introduce participants to the (Python-) code of the relevant ML-models, explain the workings of the code and interpret the outcomes from a managerial perspective. Students will be able to observe how different ML-models can be employed, how they make use of the data available to the decision maker, where they fail and where they provide useful decision support.

Within this course we make use of novel teaching formats: Each of the core sessions provides students with a presentation and accompanying video, as well as a Jupyter Notebook that allows for a (real-time) step-by-step replication and execution of the Python code that is underlying the models and individual calculations. Each of the core sessions concludes with an online assignment in which students use the Jupyter Notebooks to solve practically relevant problems.

The course is designed in such a way that students do not need prior experience in coding in Python. Our method of instruction will ensure that students understand the code and can execute the code – even without prior experience. Students who are more versed in coding and/or want to delve deeper into the implementation of the models may do so – we provide extensive supplemental material and tools for this purpose.

Gliederung:

Session 1: Introduction to Data Driven Supply Chain Management

Session 2: The Data-driven Newsvendor Problem

Session 3: The Linear Regression Newsvendor

Session 4: The Deep Learning Newsvendor

Session 5: Using Decision Trees and Random Forest

Session 6: An Outlook on Data Driven Supply Chain Management
​

Detaillierter Inhalt:

Session 1 provides students with an overview of the topic, the contents of the course, and the course mechanics. Session 1 is accompanied by step-by-step instructions on how to get started with the code environment.

Session 2 first introduces students to the case example and the data that is used throughout the entire course. It then explains the case company’s decision problem and translates it into a standard problem in Supply Chain Management (the Newsvendor Problem). Finally, first and simple approaches for solving the problem based on the company’s data are introduced.

Session 3 extends the simple approaches developed in Session 2. In a very intuitive manner, students are guided towards more elaborate approaches that allow us to leverage more of the available data, resulting in a data-driven (regression-based) solution to the Newsvendor problem.

Session 4 takes the step from conventional data-driven models (based on linear regression) to ML-based models. In this session students will learn the concept of Deep Learning and how Deep Learning models can be used to solve the company’s decision making problem.

Session 5 provides alternative ML-based approaches to solving the Newsvendor Problem. It introduces the concepts of Decision Trees and Random Forests and explains, step-by-step, how these concepts can be employed in the context of our case company’s decision making problem.

Session 6 provides insights into how the developed concepts and models can be used to tackle other decision making problems in Supply Chain Management. The session provides a number of practical examples from different industries and highlights the connection to the concepts developed in the previous sessions.

Learning Objectives:

Primary Learning Objectives:

• Develop a good general understanding of how decision making in Supply Chain Management can be improved through data-driven techniques
• Become familiar with important concepts and models in ML (Deep Learning, Random Forests, etc.) and learn how to employ such models in a Supply Chain Management context
• Be able to interpret the outcomes of data-driven models, to evaluate their performances and to identify opportunities for improvement
  
  

Secondary Learning Objectives:

• Obtain a basic understanding of how to manage and process practically relevant datasets following best practices employed by data scientists in practice
• Obtain a basic understanding of how ML models are implemented in Python and how Jupyter Notebooks can aid the work of Data Scientists
  
  

Schwierigkeitsgrad:

Einsteiger

Lehr-/Lernform:

Virtuelle Vorlesung

Interaktionsformen mit dem System/Betreuer:

Chat, E-Mail

Interaktionsformen mit Mitlernenden:

Forum

Kursdemo:

zur Kursdemo

Schlagworte:

SCM, Machine Learning, Supply Chain Management, Data-Driven Supply Chain Management, Python

Nutzung

Zielgruppe:

Ing.-Wissenschaften für FH-Studierende, Wirtschaftsinformatik für Uni-Studierende, Ing.-Wissenschaften für Uni-Studierende, Wirtschaftswissenschaften für FH-Studierende, Wirtschaftswissenschaften für Uni-Studierende, Wirtschaftsinformatik für FH-Studierende

Nutzbar im Studiengang:

Julius-Maximilians-Universität Würzburg:

• Wirtschaftswissenschaften (Bachelor)
• Wirtschaftsinformatik (Bachelor)
• Digital Business (Bachelor)

TH Würzburg-Schweinfurt:

• Wirtschaftsingenieurwesen (Bachelor)
• Business and Engieneering (Bachelor)

Otto-Friedrich-Universität Bamberg:

• Betriebswirtschaftslehre (Bachelor)
• Internationale Betriebswirtschaftslehre (Bachelor)
  
  

Geeignet für Berufsfeld:

This course prepares students for entry-level roles in supply chain analytics and data-driven decision-making. It provides a solid foundation for those interested in further studies or advancing in fields like supply chain management, data science, or operations research.

Formale Zugangsvoraussetzungen:

Course enrolment via the Virtuelle Hochschule Bayern (vhb)

Erforderliche Vorkenntnisse:

Basic math and an understanding of statistics (especially linear regression); coding skills are not required.

Erforderliche Vorkenntnisse bzgl. Handhabung der Lernplattform:

-

Verantwortlich

Trägerhochschule:

Uni Würzburg

Anbieter:

Prof. Dr. Richard Pibernik

Autoren:

Nikolai Stein, Moritz Beck, Magnus Maichle, Richard Pibernik

Betreuer:

Magnus Maichle Moritz Beck

Prüfung

Exam

Art der Prüfung:

schriftlicher Leistungsnachweis (Klausur)

Prüfer:

Prof. Dr. Richard Pibernik

Anmeldeverfahren:

Die Anmeldung erfolgt über das WWW.

Prüfungsanmeldefrist:

15.03.2025 00:00 Uhr bis 18.07.2025 23:59 Uhr

Prüfungsabmeldefrist:

15.03.2025 00:00 Uhr bis 18.07.2025 23:59 Uhr

Kapazität:

–

Prüfungsdatum:

25.07.2025

Prüfungszeitraum:

10:00 bis 11:00

Prüfungsdauer:

60 Minuten

Prüfungsort:

Uni Wuerzburg and others on request (if applicable)

Zustündiges Prüfungsamt:

Examination office of the students' home university

Zugelassene Hilfsmittel:

none

Formale Voraussetzungen für die Prüfungsteilnahme:

Registration for the exam via the vhb portal

Inhaltliche Voraussetzungen für die Prüfungsteilnahme:

Course content

Zertifikat:

Ja (graded certificate)

Anerkennung an folgenden Hochschulen:

Uni Würzburg, Uni Bamberg, FH Würzburg-Schweinfurt

Sonstige Anerkennung:

noch nicht bekannt

Online-Prüfungsan-/-abmeldung:

Ja

Bemerkung:

By completing assignments during the semester, you can collect up to 10 bonus points for the exam.

Erforderliche Technik

Browser:

Current version of a common web browser (e.g., Chrome, Edge, Firefox)

Nutzungsbedingungen

Gebühren:

Nein

Nutzungsentgelte:

für andere Personen als (reguläre) Studenten der vhb Trägerhochschulen nach Maßgabe der Benutzungs- und Entgeltordnung der vhb

Copyright:

-

Hinweise zur Nutzung:

-