After successfully completing this course, participants can design encodings for simple product configuration problems in a knowledge-based paradigm such as constraint programming or logic programming.Furthermore, participants will be able to evaluate product configurators according to criteria and features that are relevant for product configuration systems in industrial practice.

The course will be held fully online, comprising 10 meetings via videoconference (Microsoft Teams).Participants need working microphones and cameras and the willingness to switch them on during the meetings.Meetings will be recorded, recordings will be shared with the participants and deleted at the end of the semester. By registering for this course, you give your consent that your voice and video may be included in these recordings.Further recordings or screenshots of the sessions as well as any sharing of course contents with persons not participating in this course are not permitted.The meetings will be interactive, consisting of lectures, students’ presentations, discussions, and exercises.Participants will solve exercises and develop a larger project in-between meetings alone and in groups.Learning will be supported by literature, videos, software tools, and quizzes.

Product configuration is the activity of customizing a product to meet the individual needs of a particular customer. The goal is to offer such a product at a price similar to mass production and to transparently give the customer all relevant information, e.g., about environmental impact. This requires appropriate tools for specifying the product variety and recommending individual solutions.This course introduces the basics of product configuration and subsequently focuses on using Knowledge Representation and Reasoning (KRR) to implement product configuration systems.Covered topics:Product configuration, its benefits and context, scope of configurable productsModelling configuration problemsTechnologies: Variant Tables, Decision Diagrams, Feature ModelsAnswer Set Programming for Product ConfigurationConstraint Programming for Product ConfigurationGreen Configuration

Basic knowledge in logics (connectives such as and, or, implication, for all, exists) and in data modelling (esp. UML class diagrams) is recommended.

A script on the course contents will be provided to participants.Recommended additional literature:Alexander Felfernig, Lothar Hotz, Claire Bagley and Juha Tiihonen, editors. Knowledge-based Configuration: From Research to Business Cases. Morgan Kaufmann Publishers Inc., 2014Alexander Felfernig, Andreas Falkner, and David Benavides. Feature Models: AI-Driven Design, Analysis and Applications. Springer, 2024Frank van Harmelen, Vladimir Lifschitz, and Bruce Porter. Handbook of Knowledge Representation. Elsevier Science, 2007Stuart Russell and Peter Norvig. Artificial Intelligence: A modern approach. Pearson, 2022

The final grade comprises two assessment items:Homework exercises. Participants must be able to present and explain their solutions upon request after submission.Group project. Resulting documents and presentations will be graded.

The homework exercises will cover the introductory topics as well as technologies such as Variant Tables, Decision Diagrams, and Feature Models.For the group project, each group will decide whether to focus on Answer Set Programming or Constraint Programming. The project will then consist of literature research, implementation work, working with various tools, and presenting results in front of the other participants.

The homework exercises and the group project will each contribute 50% to the total grade. To pass this course you will need to attain an overall mark of 50% and achieve at least 50% in each of the assessment items.The grade will be derived from the total percentage of attained points as follows:>85% → 1>70% → 2>60% → 3>50% → 4≤50% → 5The maximum number of permitted absences, without impact on the final course grade and without having to produce an alternative piece of work, is 1 (absence with justification) resp. 0 (absence without justification).The deadline by which students can deregister without triggering an assessment is October 18.Using large language models or similar tools to solve exercises or produce text to be submitted in this course is not permitted. Using such tools to help with literature research is permitted.