User Rating: 5 / 5

Star ActiveStar ActiveStar ActiveStar ActiveStar Active

Today, I attended the China International Chemical Industry Fair (ICIF China) [中国国际化工展览会] and SpeChem [中国国际精细化工及定制化学品展览会] exhibitions in Shanghai, i.e., trade shows for the chemistry industry. ICIF is a general fair for all types of chemicals and chemistry-equipment, ranging from petroleum to fertilizer and from pumps and valves to packaging and logistics. SpeChem is focused on the special chemicals used in, e.g., the food and beverage, pharmaceutical, agricultural, electronic, water treatment, oilfield, paper, and textile industries. After already attending several logistics-related events this year, such as Intermodal and LogiMAT, I wanted to attend this fair in order to broaden my focus and to see whether IT and operations research have become central products in the chemistry industry.

The exhibition was quite large and a wide variety of products were presented. Manufacturers of organic and inorganic chemicals promoted their products, but also machinery, be it for cleaning water, automatic packaging, or producing chemicals were displayed. Hoses, valves, vessels, canisters and other utilities could be found. As a side node, I was positively surprised to find quite a lot of Anhui-based companies at the exhibition. Still, most interesting for me were producers of automation and control equipment as well as logistics providers. I even had the chance to discuss with a few of them.

The manufacturers of automation equipment for chemistry production currently seem to mainly focus on traditional features, such as fixed control procedures and networking of components, which, well, is their basic business. However, some of them seriously consider adding more intelligence to their products and production processes, be it by enhancing existing products with networked sensors, making their own production "smart," or by contemplating adding intelligent control and planning capabilities to their products. The logistics providers in the chemistry sector will need to rely on manual negotiations in the foreseeable future due to the volatility of the market and ever-changing regulations. Yet, providers for chemical freight tracking software may begin to utilize Big Data approaches to mine and provide useful information for their clients. In summary, at least from the few discussions I had, I think the chemistry industry is not yet widely utilizing the available capabilities of IT and computational intelligence, although several of my interlocutors think that this will take place in the future, will be a necessary development, and/or are seriously planning to move into this direction.

I believe that the chemistry industry could largely profit from a movement similar to Industry 4.0 (may I dare to call it Chemistry 4.0?). Often, they deal with automated processes run by huge machinery. Questions such as the automated configuration of their process parameters and the best scheduling of production tasks and maintenance works arise quite naturally. Also, chemical manufacturers often do not produce based on customer orders but perform stockpiling, which asks for intelligent predictions of future orders. All of the above could be single optimization problems or even combined into a complex integrated optimization task, maybe even involving optimization along the whole supply chain. Of course, to get there means to first invest money into the development of optimization tools, without clearly knowing how much resources or money can be saved or how much the productivity can be increased beforehand. This may be an understandable and large hurdle, nevertheless, I believe that there is a huge potential in this field.

User Rating: 3 / 5

Star ActiveStar ActiveStar ActiveStar InactiveStar Inactive

On August 3, 2017, I presented my research talk Automating Scientific Research in Optimization at the Database Systems and Data Mining Group of Prof. Thomas Seidl at the Institute of Informatics an the Computational Statistics Group of Prof. Dr. Bernd Bischl who is at the Institute of Statistics, both of the Faculty of Mathematics, Informatics, and Statistics of the Ludwig-Maximilians-Universität Munich (LMU) in Munich (München), Germany. The talk was organized by Dr. Markus Mauder.

The Database Systems and Data Mining Group conducts research on data science, data mining, machine learning, artificial intelligence and database technologies. They conduct both fundamental and applied research and aim to supporting partners from many domains such engineering, business, humanities and life sciences in analyzing huge and complex data. Their research topics include explorative data analytics (clustering, outlier detection and interactive data mining), new computation models for data mining (cloud architectures, MapReduce, deep learning, GPU computing, embedded analytics), data mining for complex objects (spatio-temporal data, graphs and networks, multimedia, community detection, trajectory analysis, network queries, pattern mining, dealing with uncertain data,  as well as games and sports analytics), process mining (temporal aspects in process models, event identification, process model identification, stream process mining), and similarity search. Since my presentation is mainly on applying data mining and machine learning to extract information from experiments with optimization methods, this provided good grounds for nice discussions after the presentation. They collaborate with partners like Siemens and BMW.

The Computational Statistics Group conducts research in predictive modelling (e.g., non-linear classification and regression), ensemble techniques, statistical and machine learning, statistical computing and computational statistics, model- and variable selection, expensive black-box optimization, bayesian optimization, and statistical software development. They organize the Munich City R courses, R being the one of the most important statistics software packages out there right now and is also used inside of our optimizationBenchmark software. They have contributed a huge set of awesome software packages, including, for instance, a machine learning toolbox for R and an R package for the COCO framework (one of the works related to our talk).

Prof. Bischl is furthermore one of the founders of COSEAL, the COnfiguration and SElection of ALgorithms group, an international group of researchers with focus on algorithm selection and algorithm configuration. These two topics are the exact complement of what I presented in my talk: While my system analyzes how algorithms behave and tries to find reasons (which a researcher can use to improve the algorithm), the work of COSEAL is trying to find ways to automatically find/choose the best behaving algorithms (or algorithm configuration). The group is very active and organizes workshops, competitions, and provides software and benchmark datasets on their fields. (In September 2017, I became a member of COSEAL.)

Giving a presentation at LMU with attendants from both groups was very exciting for me. I came here with the goal to broaden my horizon and to learn where these experts see merits and errors in my presented approach. The after the presentation discussion was indeed very interesting and fruitful.

User Rating: 3 / 5

Star ActiveStar ActiveStar ActiveStar InactiveStar Inactive

On August 2, 2017, I gave my research talk Automating Scientific Research in Optimization at the Algorithmics Group of Prof. Dr. Ernst Althaus at Institute of Computer Science of the Johannes Gutenberg University Mainz (JGU, Johannes Gutenberg Universität) in Mainz, Germany.

The Algorithmics group is mainly known for their contributions to algorithm engineering for combinatorial optimization problems. They focus on the development and implementation of algorithms for interdisciplinary problems, which require both the theoretical development of algorithms as well as their practical implementation. Their application areas are real-time scheduling (as needed in many embedded systems), the verification of discrete-continuous hybrid systems (again needed in embedded systems design), and dense sub-graphs of Steiner tree problems that arise in chip design, bioinformatics, social network analysis, and other areas of research and which they tackle with mixed integer linear programming with cutting planes and dynamic programming. Additionally, they publish several works on genetics-related topics such as sequence alignment problems.

Meeting this group and presenting at the JGU was a very nice experience and I am thankful that the talk was visited well, even though it was in the holiday season.

User Rating: 3 / 5

Star ActiveStar ActiveStar ActiveStar InactiveStar Inactive

On August 1, 2017, I gave the research presentation Automating Scientific Research in Optimization at the research group of Prof. Dr. Rainer Schrader at the Department of Computer Science of the University of Cologne (Universität zu Köln).

This group pursues the goal of building a bridge between real-life problems and the theoretical theorems of pure Mathematics. They closely cooperate with industrial partners to obtain solutions in an application-driven way, by applying research in several fields of Mathematics and Computer Science. Their current work includes contributions to the theory of interval graphs. The knowledge transfer to the industry plays also a very important role for the group as well, as it inspires innovative directions for both their fundamental research and teaching curriculum. The group offers research strength in areas such as complex simulations for economical models and traffic, the development of control models for optimizing process in industrial production, pattern recognition, and the development of multimedia and interactive learning systems. These competences support a variety of partners such as DB Schenker, the Deutsche Bank AG, the Deutsches Zentrum für Luft- und Raumfahrt (DLR), Ford, Lufthansa Systems GmbH, the Siemens AG, the Springer-Verlag, and the WDR TV sender (ARD Sportschau). The fundamental research of the group concerns discrete structures such as partially ordered sets and P4 structures, combinatorial optimization, bioinformatics, and stochastic modeling (such as the aforementioned traffic models).

I was looking forward to meeting Prof. Schrader and his group for quite some time and found the atmosphere in their Tuesday's Research Seminar very enjoyable.

User Rating: 3 / 5

Star ActiveStar ActiveStar ActiveStar InactiveStar Inactive

On July 31, 2017, I held my research talk Automating Scientific Research in Optimization at Swarm Intelligence and Complex Systems Group of Prof. Dr. Martin Middendorf at the Department of Computer Science, Faculty of Mathematics and Computer Science of the University of Leipzig in Leipzig, Germany (official announcement).

The group is focused on the field of complex systems which they tackle with swarm intelligence, bio-inspired computation and optimization, and organic computing. The also tackle issues in phylogeny and reconfigurable architectures for parallel computing. One of the many contributions of this group is a large set of open source software systems, for phylogenesis, RNA and mitochondrial genome analysis using optimization algorithms. I know Prof. Middendorf for more than five years personally and have deep respect for his work: His Population-based Ant Colony Optimization algorithm is the best (non-hybrid) metaheuristic approach that I know for the Traveling Salesman Problem and also a component of the best hybrid approach in all the experiments I have conducted with our TSP Suite in, e.g., our 2014 study.

After I met Prof. Middendorf several times in China and know him since 2011, so it was very nice to finally get to visit his group in Leipzig. It was a really pleasant experience to present in his faculty and I enjoyed the interest and fruitful discussion afterwards, as well as Prof. Middendorf's hospitality.

feed-image rss feed-image atom