Universität Wien

040676 KU Metaheuristics (MA) (2020S)

4.00 ECTS (2.00 SWS), SPL 4 - Wirtschaftswissenschaften
Prüfungsimmanente Lehrveranstaltung

An/Abmeldung

Hinweis: Ihr Anmeldezeitpunkt innerhalb der Frist hat keine Auswirkungen auf die Platzvergabe (kein "first come, first served").

Details

max. 30 Teilnehmer*innen
Sprache: Englisch

Lehrende

Termine (iCal) - nächster Termin ist mit N markiert

Montag 20.04. 15:00 - 18:15 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Montag 20.04. 18:30 - 20:00 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 21.04. 13:15 - 14:45 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 21.04. 15:00 - 16:30 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 21.04. 16:45 - 18:15 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 21.04. 18:30 - 20:00 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Mittwoch 22.04. 13:15 - 16:30 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Montag 04.05. 13:15 - 14:45 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Montag 04.05. 15:00 - 18:15 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Montag 04.05. 18:30 - 20:00 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 05.05. 13:15 - 18:15 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Dienstag 05.05. 18:30 - 20:00 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Mittwoch 06.05. 13:15 - 14:45 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Mittwoch 06.05. 15:00 - 18:15 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß
Mittwoch 06.05. 18:30 - 19:50 PC-Seminarraum 1 Oskar-Morgenstern-Platz 1 1.Untergeschoß

Information

Ziele, Inhalte und Methode der Lehrveranstaltung

Heuristics form an indispensable tool for everyone working in operations management because problems arising from practice are often too hard to solve exactly and heuristics are relatively simple methods that may provide feasible solutions of good quality. The course aims at introducing the main concepts of heuristics solution methods and of the modern metaheuristics paradigms developed in the last decades. The methods will be illustrated on the to solve general problems but uses logistics optimization problems such as the vehicle routing problem as examples of specific implementations. The field of routing is in fact so rich that virtually all published heuristic ideas have been applied to it.

The course is further divided into two parts, each of which first covers general problems and then focuses on routing:
• Classical heuristics to construct a feasible solution and improvement heuristics based on structured local search
• Metaheuristics aiming at escaping local optima
All methods will be illustrated through actual implementations in high level language.
By the end of the course students will be able to:
- Identify problems which will require heuristics for their solution.
- Identify the appropriate heuristic for the solution of a problem.
- Discuss the characteristics of the main heurisitc and metaheuristic techniques by considering their efficiency and effectiveness.
- Develop implementations of heuristics for optimization and vehicle routing problems.

Art der Leistungskontrolle und erlaubte Hilfsmittel

The Assessment of the course will be provided by a written exam covering both methodological open questions and simple exercises on the application and modification of heuristics. No programming is included in the exam.
Given the methodological content of the exam, lecture notes and other books are not permitted during the exam.
During the course will be discussed with the students the possibility of having a homework assignment. In case the homework assignment is defined it will amount 20% of the final grade and the written exam 80%, otherwise the written exam will account for 100% of the final grade.

Mindestanforderungen und Beurteilungsmaßstab

Appropriate points will be assigned to each part of the exam and to the possible homework, the grading will be scaled in 100%.
In order to pass the course (minimum requirement) students have to achieve at least 50% in total.

The other grades are distributed as follows:
4: 50% to <63%
3: 63% to <75%
2: 75% to <87%
1: 87% to 100%

Prüfungsstoff

Greedy and construction heuristics
Local search improvement methods
Metaheuristics (evolutionary and local search based)

Literatur

Lecture notes and coding companion available online.

Selection of chapters from these books:

· Michalewicz, Z. and Fogel, D.B. (2004). How to solve it:
modern heuristics. Springer

· Talbi, El-Ghazali (2009). Metaheuristics: From Design to
Implementation. Wiley

· Toth, P. and Vigo, D. (2002). The Vehicle Routing Problem, 1st
edition. SIAM.

· Toth, P. and Vigo, D. (2014). Vehicle Routing: Problems,
Methods and Applications, 2nd edition. SIAM.

Zuordnung im Vorlesungsverzeichnis

Letzte Änderung: Mo 07.09.2020 15:19