Prof. Dr. Ivo Nowak

Fakultät Nachhaltige Ingenieurwissenschaften
Professor

Berliner Tor 21 ,
20099 Hamburg

Raum 410

+49 40 428 75-8789
ivo.nowak (at) haw-hamburg (dot) de

Programm CEADS Computational Engineering and Applied Data Science
Doktorandenbetreuer

ivo.nowak (at) haw-hamburg (dot) de

Programm CEADS Computational Engineering and Applied Data Science
Sprecher eines Promotionsprogramms

ivo.nowak (at) haw-hamburg (dot) de

Lehrgebiete / Tätigkeitsfelder

................................................................................................................................................................................................

Schwerpunktthemen / Kernkompetenzen

................................................................................................................................................................................................

Kurzbiographie

................................................................................................................................................................................................

Sonstige Informationen

................................................................................................................................................................................................

Research

Research Group Machine Learning

Current Projects

iMath "An Intelligent System to Learn Mathematics" (2022 - 2024). The iMath project is funded, by the European Commission through the Portuguese National Agency for the Erasmus+ Programme to develop a new AI-driven tool that supports higher education math students, providing them with a database of resources, hands-on activities, self-evaluation tests based on their previous performances, see iMath - project and iMath - brochure (English and German)
LD-SODA "Learning-based data analysis - stochastics, optimization, dynamics and approximation" (2020-2023, Landesforschungsförderung Hamburg), https://www.math.uni-hamburg.de/home/iske/soda.en.html
In order to reduce risks when using machine learning methods and to improve existing algorithms, an understanding of the underlying mathematical methods of data analysis and machine learning is essential. In cooperation with Sarah Hallerberg (HAW), Armin Iske (UHH) and Mathias Trabs (UHH), we are addressing relevant mathematical questions of machine learning from the involved disciplines stochastics, optimization, dynamics and approximation.
DADLN "Dynamics and adaptive decomposition of machine learning networks", (2020-2023, BMBF), AI research program
SwarmPicker(2020-2023): In cooperation with H.-J. Schelberg, S. Hallerberg, S. Schulz and J. Dahlkemper we are developing of a swarm of autonomous service robots for cleaning inhomogeneous environments.
OEB-EnSys "MINLP-Optimization of Design and Operation of Complex Energy Systems" (2017-2020, BMWi)
DECOGO "Decomposition Methods for MINLP" : The goals of this project are (i) development of new parallel decomposition methods for deterministic global optimization (ii) development of the new open-source MINLP-solver Decogo and (iii) solving difficult industrial optimization models using the new methods.

Large-Scale Global Optimization

Motivated by Column Generation methods for solving huge optimization problems with Millions of variables, we develop new decomposition methods for deterministic global optimization, integrate these methods into the new parallel MINLP-solver Decogo, and solve difficult industrial optimization models.
P. Muts, S. Bruche, I. Nowak, O. Wu, E. Hendrix, G. Tsatsaronis, A Column Generation Algorithm for Solving Energy System Planning Problems, Optimization and Engineering, 2021
P.Muts, Decomposition Methods for Mixed-Integer Nonlinear Programming, PhD thesis, 2021
P. Muts, I. Nowak, Eligius M.T. Hendrix, On decomposition and multiobjective-based column and disjunctive cut generation for MINLP, Optimization and Engineering, link.springer.com/article/10.1007/s11081-020-09576-x , 2020
P. Muts, I. Nowak, Eligius M.T. Hendrix, A Resource Constraint Approach for One Global Constraint MINLP, in Computational Science and Its Applications - ICCSA 2020, see link.springer.com/chapter/10.1007%2F978-3-030-58808-3_43
P. Muts, I. Nowak, Eligius M.T. Hendrix, The Decomposition-based Outer Approximation Algorithm for Convex Mixed-Integer Nonlinear Programming, Journal of Global Optimization, doi.org/10.1007/s10898-020-00888-x, 2020
I. Nowak, P. Muts, Combining Column Generation and Outer-Approximation for Solving Nonconvex MINLPs, presentation, 2019
P. Muts, I. Nowak, Towards Multi-Tree Methods for Large-Scale Global Optimization, Proceedings of the World Congress of Global Optimization, 2019
I. Nowak, P. Muts, Decomposition-based Successive Approximation Methods for Global Optimization, Proceedings of LEGO, 2018
I. Nowak, N. Breitfeld, E. M. T. Hendrix, G. Njacheun-Njanzoua, Decomposition-based Inner- and Outer-Refinement Algorithms for Global Optimization, Journal of Global Optimization, 2018
I. Nowak, P. Muts, Decomposition-based Successive Approximation Methods for Global Optimization, presentation, 2018
I. Nowak, Parallel Decomposition Methods for Nonconvex Optimization - Recent Advances and New Directions, Proceedings of MAGO, 2014
I. Nowak and S. Vigerske, LaGO - a (heuristic) Branch and Cut algorithm for nonconvex MINLPs, Central European Journal of Operations Research 16:2, 127-138, 2008
I. Nowak and S. Vigerske, LaGO - MINLP Solver: github.com/coin-or/LaGO, LaGO presentation, 2005
I. Nowak, Relaxation and Decomposition Methods for Mixed Integer Nonlinear Programming, Basel, International Series of Numerical Mathematics, Vol. 152, XVI, 213 p., 2005, MINLP book

Machine Learning and Decision Support

We develop methods for machine learning and for making decisions under uncertainty, e.g. in revenue management, aircraft scheduling, or energy system planning
N. Yalcin, Possible applications and potentials of artificial intelligence in the product development process, master thesis, HAW Hamburg, 2020
S. Lang, Development of a simulation environment to investigate the influence of predictive maintenance on the operational operation of an aircraft fleet, master thesis, HAW Hamburg, 2017
T. Winter, I. Nowak, J. Pancake-Steeg, S. Würll, A Frank-Wolfe Decomposition Approach to Network Revenue Management , presentation, OR2017 Berlin, Germany, 2017
R. Borndörfer, I. Dovica, I. Nowak, T. Schickinger, Robust Tail Assignment, working paper, Anna Valicek Medal Finalist, 2010
I. Nowak and S. Vigerske, Adaptive discretization of convex multistage stochastic programs, Mathematical Methods of Operations Research, 65:2, 2007

Planning and Control

We develop methods for solving planning and control problems, like in supply chain, manufacturing, energy system planning or airline management.
I. Nowak, P. Muts, E. M.T. Hendrix, Multi-Tree Decomposition Methods for Large-Scale Mixed Integer Nonlinear Optimization, in book ‘Large Scale Optimization in Supply Chain & Smart Manufacturing: Theory & Application’, Editors: J. Velásquez-Bermúdez, M. Khakifirooz, M. Fathi, Springer Optimization and Its Applications, 2019
L. Krohn, Smart Heat Grid Hamburg: Development of a sustainable heat supply concept for a industrial and residential area as nucleus for perspectively connection to an existing district heating grid, master thesis, HAW-Hamburg, 2019
T. Ahadi-Oskui, S. Vigerske, I. Nowak, G. Tsatsaronis, Optimizing the design of complex energy conversion systems by Branch and Cut, Computers & Chemical Engineering 34:8, 1226-1236, 2010
I. Nowak, V. Gintner, A Dynamic Reduce and Generate Approach for Airline Crew Scheduling, Column Generation 2008, June 17-20, Aussois, France

Engineering Design

We develop methods for solving engineering design and control problems, like structural and topology optimization problems
M. Schelle, Combined trajectory and footstep planing for quadruped robots using linear model predictive control, master thesis, HAW-Hamburg, 2020 (Franz-Herbert Spitz award for best master thesis)
Jascha Joujan, Development of an algorithm to optimize the layer thickness distribution of heating lacquer systems using the finite-element-method, master project, HAW-Hamburg, 2020
M. Schelle, Optimization of the driving strategy for a racing car using optimal Control, master project, HAW-Hamburg, 2020
T. Duffe, Concept layout and numerical development of a structural roll formed crash element for commercial vehicles by utilizing finite element analysis, master thesis, HAW-Hamburg, 2019
M. Paulenz, Investigation of integration of parameter-based design optimization of casting parts in the product development process, master thesis, HAW-Hamburg, 2018
S. Ehlers, Experimente mit einem Dekompositionsverfahren zur Topologieoptimierung, master project, HAW-Hamburg, 2018
M. Paulenz, Numerische Experimente mit einer Multistart-Methode zur Topologieoptimierung und Anbindung an Hyperworks/Optistruct, master project, HAW-Hamburg, 2018
G. Njacheun-Njanzoua, I. Nowak, Solving MIQQP Topology Optimization Problems by Successive Nonconvex Approximation, presentation, OR2017 Berlin, Germany, 2017
M. Tran, Simulation-based determination of friction coefficients on self-pierce riveting with semi-hollow rivets, master thesis, HAW-Hamburg, 2016
K. Müller, Globale Topologieoptimierung, master project, HAW-Hamburg, 2015
I. Nowak, Minimal Surfaces with Prescribed Topological Type on a Schwarzian Chain in M3(c), Annals of Global Analysis and Geometry 11 : 331-344, 1993

For more publicatios see Research Gate