040684 PR KFK PM/SCM/TL: Praktikum Metaheuristics II (SCM) (2008W)
Prüfungsimmanente Lehrveranstaltung
Labels
Block LV
11.11, 25.11,09.12.2008 09:00-14:00 Uhr Lab 6
13.01,27.01.2008 09:00-14:00 Uhr Lab 6
11.11, 25.11,09.12.2008 09:00-14:00 Uhr Lab 6
13.01,27.01.2008 09:00-14:00 Uhr Lab 6
An/Abmeldung
Hinweis: Ihr Anmeldezeitpunkt innerhalb der Frist hat keine Auswirkungen auf die Platzvergabe (kein "first come, first served").
- Anmeldung von Mo 01.09.2008 09:00 bis So 21.09.2008 23:59
- Anmeldung von Fr 26.09.2008 09:00 bis Fr 05.12.2008 23:59
- Abmeldung bis Di 14.10.2008 23:59
Details
max. 30 Teilnehmer*innen
Sprache: Englisch
Lehrende
Termine (iCal) - nächster Termin ist mit N markiert
- Dienstag 11.11. 09:00 - 14:00 EDV-Labor 6
- Dienstag 25.11. 09:00 - 14:00 EDV-Labor 6
- Dienstag 09.12. 09:00 - 14:00 EDV-Labor 6
- Dienstag 13.01. 09:00 - 14:00 EDV-Labor 6
- Dienstag 27.01. 09:00 - 14:00 EDV-Labor 6
Information
Ziele, Inhalte und Methode der Lehrveranstaltung
Art der Leistungskontrolle und erlaubte Hilfsmittel
Weekly home assignments (30%), term project (40%), final test (30%)
Mindestanforderungen und Beurteilungsmaßstab
The goal of the course is that students build up an intuition whether and why a certain metaheuristic solves a certain planning problem from SCM well. Therefore, they will learn how to design efficient metaheuristics on their own. Students shall understand the impact of core design decisions which must be made for any metaheuristic: (1) developing a useful solution representation and (2) linking it to the choice of search operators, (3) evaluating the quality of solutions and (3) making sensible choices for initial solutions.
Through their own experience, discussion in class and working with the literature, students will build up a deep understanding of how metaheuristics for strategic network design, auctions and multi-echelon inventory problems should be designed in order to work well.
Through their own experience, discussion in class and working with the literature, students will build up a deep understanding of how metaheuristics for strategic network design, auctions and multi-echelon inventory problems should be designed in order to work well.
Prüfungsstoff
The course will consist of focused lectures on the SCM problems, as well as focused lectures on the design and application of metaheuristics. Students will work on exercises, cases and implementation tasks. Therefore, basic programming skills in a high-level programming language such as C, C++, Java, VB or other are required. A basic understanding of linear programming is helpful but not required.
Literatur
A literature list will be distributed after the start of the course.
Zuordnung im Vorlesungsverzeichnis
Letzte Änderung: Mo 07.09.2020 15:29
These problems are often hard to solve with exact planning methods. Thus, metaheuristics are widely applied in practice to reliably generate high quality solutions in reasonable time. We will study various metaheuristic solution approaches for these problems, figure out why and how they work well, and how they can be adapted to new problems which have not been solved before. This will contain some implementation effort, so that basic programming skills are helpful.