Meinolf Sellmann is the most recent junior faculty member to receive a National Science Foundation (NSF) CAREER grant for his Cornflower project proposal.
The dawn of the new century casts light on three dramatic economic challenges that will determine our future as a society: demography, globalization, and shortage of natural resources. Today we need to develop the technology that will allow our children to maintain our standard of living. Therefore, we need to find ways to increase our economical efficiency. Computer science can play a decisive role when facing this challenge. After more than 60 years of research, algorithmic computer science can offer a lot to help saving. However, the main impact of computational optimization support is currently limited to large companies in few core application areas such as the transportation industry. Specialized algorithmic solutions have shown to be extremely successful in these domains. While the algorithmic technologies that were developed are by no means specific to the current areas of application, the main obstacle for a broader realization of their vast potential is largely due to the lacking ease of use. Therefore, the main goal of the Cornflower Project is to develop techniques that allow inexperienced users to exploit optimization power efficiently.
Optimization techniques have been developed in three historically separate research areas: constraint programming, operations research, and algorithm theory. While the main focus of the project is to provide a high level of automization and algorithms that can be hooked to intuitive modeling primitives that facilitate the use of intelligent optimization support, we do not stop at the boundaries of traditional research areas. Instead, we integrate and hybridize ideas developed in different communities in order to provide easily accessible high performance optimization technology. Particularly, we focus on the development of high-level constraints that allow users to model their problems as conjunctions of intuitive substructures and provide hybrid methods for their efficient combination. Moreover, we develop automization techniques for the handling of symmetries that can be the cause of severe inefficiencies when handled poorly.
The main goal of the Cornflower project is to solve the algorithmic problems that arise in the context of optimization driven decision support systems that are intuitive to use and that provide a high level of automization. By making algorithms and methods publicly available in the Cornflower Library, the project contributes to widening the access to computational decision support.
Our goal to broaden the accessibility of computational decision support while breaking the barriers between theoretical and practical computer science goes hand in hand with the educational goals of the project. Optimization techniques play an ever more important role for many other CS disciplines. By developing optimization courses that integrate methods from various research areas, Cornflower helps to educate the next generation of experts in combinatorial algorithms, while students moving on to other disciplines take away a sound understanding how to tackle combinatorial problems and how to utilize standard solvers for their own purposes. Attracting both students who are mainly interested in technical methodology as well as those who are primarily attracted to specific applications is also a promising approach to broadening participation in science at large. Through the outreach program Artemis, a five-week summer program for high-school girls, young are encouraged women to pursue careers in science and engineering.
NSF’s CAREER program recognizes and supports the early career development activities of those faculty members who are most likely to become the academic leaders of the 21st century. CAREER awardees will be selected on the basis of creative, integrative, and effective research and education career development plans that build a firm foundation for a lifetime of integrated contributions to research and education.