TUM – TUM – Menü

 

 
SCoNDO II

Students' Conference on Nonlinear and Discrete Optimization

 

Acknowledgements

Organizers

The conference is organized by
Melanie Bestle
Florian Lindemann
Boris von Loesch
MichaelRitter

Conference Program

Morning Sessions

The morning program consists of two sessions with two students' talks scheduled per session. Each talk will be 30 minutes plus discussion. There is a coffee break in between the two sessions.

9:00 a.m. conference opening
9:15 a.m. talk Optimizing Container Usage in the Logging Industry
10:00 a.m. talk Topology Optimization - get an optimal layout!
11:15 a.m. talk Robust Warehousing
12:00 p.m. talk Scheduling and Routing in Home Healthcare

Lunch break from 12:45 p.m. until 2:00 p.m.

Afternoon Sessions

There are two afternoon sessions featuring a total of four students' talks (30 minutes each, plus discussion) and an invited talk (30 minutes, including discussion). There is a coffee break between the two sessions. Additionally, we will have a closing session for evaluation and presentation of certificates.

2:00 p.m. talk The Sightseeing Problem - A smart way to plan your vacation!
2:45 p.m. invited talk Quadratic optimization problems relevant to the Erdős bipartification conjecture
3:15 p.m. talk Bridge maintenance planning in Munich
4:30 p.m. talk Multimodal Image Registration
5:15 p.m. talk The New Age of Power Supply
6:00 p.m. evaluation, discussion and certificates

Conference Dinner

We meet for conference dinner at 7:00 p.m. at "Ristorante Il Salento" , Heßstraße 15, where some tables are reserved for the conference dinner. The restaurant is within about 10 minutes walking distance from the conference venue.


Größere Karte anzeigen

Conference Venue

The conference takes place at room 0606 at
Technische Universität München, Stammgelände
main building
Arcisstraße 21

Please remember to enter through the main entrance using your student ID if necessary (or stating that you want to visit the conference or the university library) as the other entrances will be closed on Saturday.

Parking is very limited in the area, thus public transport is recommended for getting to the conference venue. The closest subway stops are Theresienstraße (U2), Königsplatz (U2) and Odeonsplatz (U3/4/5/6), the closest bus stop is "Technische Universität" (bus 100).

Abstracts and Speakers

Topology Optimization - get an optimal layout!

Moritz Keuthen, Stephanie Nikola
Topology Optimization is used to find an optimal structure with respect to attributes like a maximum of stiffness or a minimum of weight under certain conditions. The main focus of our work was the creation of cable car pillar which has maximal stiffness under restricted volume. To achieve this we created a suitable mathematic model, implemented necessary extensions to an existing Topology Optimization code and conducted a series of tests.

Robust Warehousing

Christine Abe, Stephanie Troppmann, Andrej Winokurow
Robust optimization takes the uncertainty of several parameters into account to determine a solution, which is robust. This means the robust solution satisfies every constraint for all realities of the input parameters. We applied this approach to an optimization problem in a warehouse, as in this environment are lots of uncertain factors, e.g. the daily demand of products.

Scheduling and Routing in Home Health Care

Jan Erik Müller, Patricia Rachinger, Wolfgang Riedl, Tina Janne Schmidt, Katharina Zahnweh
The demand for home health care services, which ranges from simple household assistance tasks to medical treatment, is rising tremendously. Thus, optimal scheduling is desired. Our aim is to minimize the driving and waiting times of the nurses as well as to maximize the satisfaction level of clients and nurses. First, we have modeled this generalization of the vehicle routing problem with time windows as a 2-index formulation. Using this formulation, we found a gap in complexity, which prevented us from solving instances with more than four nurses and 20 clients. Therefore, an approach using Column Generation on the set of all feasible routes and a TSP formulation that is based on a multicommodity flow model were implemented in Mosel. In our talk, we will discuss these three different formulations and how they could be combined to achieve better running times.

The Sightseeing Problem - A smart way to plan your vacation!

Martin Huber, Beate Müller, Katia Nédélec, Stephanie Nikola, Thomas Schmelz, Daniela Steidl
In the Sightseeing Problem, you try to find an optimal tour of sights in a certain city. Optimality is thereby measured by user-defined weights of interest per sight. Furthermore constraints on the duration of the tour as well as on the cost of the tour must be considered. To solve this problem, there are two different approaches: On the one hand you can approximate the solution by a heuristic, on the other you can search for an optimal tour by using an adopted branch & cut algorithm.

In our work, we added a new constraint, called "must-nodes" to the system. Thereby the tourist can specify which sights must be included in the tour. Our work was based on an existing implementation, in which we extended both the heuristic and the branch\&cut algorithm to deal with must-nodes.

Quadratic optimization problems relevant to the Erdős bipartification conjecture

Peter Heinig
A conjecture of Erdős, published for more than thirty years (yet open to this day) asks whether every simple undirected and triangle-free graph with n vertices can be made into a bipartite graph by deleting at most 1/25 n^2 edges. Two years ago, building on work of Brandt, Chen, Jin and Koh, I found a way to make some new inroads toward a special case of the conjecture. This leads to some quadratic optimization problems with a dihedral group acting transitively on the variables. Up until now, I could not carry out what I had planned to do and have long since turned to other problems, but in this talk I will give a short description of what remains to be done for my approach.

Bridge maintenance planning in Munich

Andrea Elbrächter, Lucia Fogelstaller, Fabian Klemm, Sarah Nieswand, Leonhard Schiele
Munich is a beautiful city with a large number of bridges which go over the river Isar. These bridges have to be maintained from time to time, so it is interesting to find the best point in time to do this work. And that’s exactly our aim - to create an optimal maintenance plan for the next 15 years. We also have to deal with certain constraints like assuring the accessibility of the districts, complying with street and personnel capacities and observing the deadlines of the bridges. You will get to know our basic model and besides this you will learn how we find the Nash equilibrium in a Min-Cost-Flow problem.

Multimodal Image Registration

Nicole Bengesser, Tobias Böttger, Lukas Karge, Andrei Morozyuk
Multimodal Image Registration is a process often used in medical diagnostics, which aligns two or more images of different modalities. This makes it possible to integrate the useful information from the separate images. In our project we compared different similarity measures in combination with different transformations and optimization algorithms to develop an effective procedure.

The New Age of Power Supply

Julia Bergbauer, Anna Sanden, Pascal Schambach, Andrej Winokurow
The rise of renewable energies leads to high fluctuations in electric power generation. The questions is how can - at the same time - the consumer demand for electricity be matched by the providers?

Our objective was to include for the first time geographical realities and power line restrictions into the model. We present the most efficient configuration of power generation and hydroelectricity storage calculated for Germany.

Research Unit M9


Department of Mathematics
Boltzmannstraße 3
85748 Garching b. München
Germany
phone:+49 89 289-16858
fax:+49 089 289-16859
sekretariat-m9ma.tum.de

Professors

Prof. Dr. Peter Gritzmann
Applied Geometry and Discrete Mathematics

Prof. Dr. Andreas S. Schulz
Mathematics of Operations Research
(affiliated member of M9)

Prof. Dr. Stefan Weltge
Discrete Mathematics

News

April 2018
Case Studies 2018: Save the date: Case Studies poster presentation on May 25th, 2018, final workshop on July 7th, 2018.