1. the complete title of one (or more) paper(s) published in the open literature describing the work that the author claims describes a human-competitive result; i): Automatic innovative truss design using grammatical evolution ii): Discrete Planar Truss Optimization by Node Position Variation using Grammatical Evolution ---------------------------------- ---------------------------------- 2. the name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s); Michael Fenton, UCD NCRA, Block D, UCD Michael Smurfit Graduate School of Business, Carysfort Avenue, Blackrock, Co. Dublin, Ireland. michaelfenton1@gmail.com +353 86 326 5665 Ciaran McNally, UCD School Of Civil Engineering, Richview Newstead, Belfield, Dublin 4, Ireland. ciaran.mcnally@ucd.ie +353 1 716 3202 Jonathan Byrne, Kilmoon Cross Nurseries, Ashbourne, County Meath, Ireland. jonathan.byrne@intel.com +353 86 325 7989 Erik Hemberg, 32 Vassar St, Cambridge, MA 02139, USA. erik.hemberg@gmail.com +1 617 335 3488 James McDermott, UCD NCRA, Block D, UCD Michael Smurfit Graduate School of Business, Carysfort Avenue, Blackrock, Co. Dublin, Ireland. james.mcdermott2@ucd.ie +353 1 716 8031 Michael O'Neill, UCD NCRA, Block D, UCD Michael Smurfit Graduate School of Business, Carysfort Avenue, Blackrock, Co. Dublin, Ireland. m.oneill@ucd.ie +353 1 716 8048 ---------------------------------- ---------------------------------- 3. the name of the corresponding author (i.e., the author to whom notices will be sent concerning the competition); Michael Fenton ---------------------------------- ---------------------------------- 4. the abstract of the paper(s); i): Automation in Construction Truss optimization in the field of Structural Engineering is a growing discipline. The application of Grammatical Evolution, a grammar-based form of Genetic Programming (GP), has shown that it is capable of generating innovative engineering designs. Existing truss optimization methods in GP focus primarily on optimizing global topology. The standardmethod is to explore the search spacewhile seeking minimumcross-sectional areas for all elements. In doing so, critical knowledge of section geometry and orientation is omitted, leading to inaccurate stress calculations and structures not meeting codes of practice. This can be addressed by constraining the optimisation method to only use standard construction elements. The aim of this paper is not to find fully optimized solutions, but rather to show that solutions very close to the theoretical optimum can be achieved using real-world elements. This methodology can be applied to any structural engineering design which can be generated by a grammar. ii): IEEE Transactions on Evolutionary Computation The majority of existing discrete truss optimization methods focus primarily on optimizing global truss topology using a ground structure approach, in which all possible node and beam locations are specified a priori. The ground structure discrete optimization method has been shown to be restrictive as it limits derivable solutions to what is explicitly defined. Greater representational freedom can improve performance. In this paper Grammatical Evolution is applied. It can represent a variable number of nodes and their locations on a continuum. A novel method of connecting evolved nodes using a Delaunay triangulation algorithm shows that fully triangulated, kinematically stable structures can be generated. Discrete beamtruss structures can be optimized without the need for any information about the desired form of the solution other than the design envelope. Our technique is compared to existing discrete optimization techniques, and notable savings in structure selfweight are demonstrated. In particular our new method can produce results superior to those reported in the literature in cases where the problem is ill-defined and the structure of the solution is not known a priori. ---------------------------------- ---------------------------------- 5. a list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria (see above) that the author claims that the work satisfies; B, E, G ---------------------------------- ---------------------------------- 6. a statement stating why the result satisfies the criteria that the contestant claims (see examples of statements of human-competitiveness as a guide to aid in constructing this part of the submission); The methods and results presented in these papers consist of a novel method for automatically designing engineering truss structures comprised of discrete elements. The method is based on Grammatical Evolution, a grammar-based form of Genetic Programming. ---------------------------------- B ("The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal.") Our paper in the IEEE Transactions on Evolutionary Computation describes a novel method to derive solutions that are either equal to or better than the best scientific results from the literature at the time of publication for a number of benchmark problems. The first two problems had six comparisons from peer-reviewed scientific literature [14, 43, 44, 45, 46, 47] and one earlier iteration of our own algorithm (paper i). In both cases, results from either paper (i) or (ii) out-performed all comparisons from the literature. The third problem had three comparisons from peer-reviewed scientific literature [45, 48, 49] and one earlier iteration of our own algorithm (paper i). In this case the results from paper (ii) were able to exactly match the best performance found in the literature. These results are significant, as our approach significantly increases the difficulty of the problems attempted by the literature by adding numerous real-world constraints, including: - Constraints in the form of building standards and codes of practice, - Limiting the available construction materials to a discrete set of readily available common construction elements from steel suppliers. Furthermore, we have changed the respresentation of the problem to greatly increase the degree of freedom with which problems can be generated. The combination of these changes to the problem means that any results that can match or improve on the performance of the best solutions from the literature are most impressive. ---------------------------------- E ("The result is equal to or better than the most recent human-created solution to a long-standing problem for which there has been a succession of increasingly better human-created solutions.") The literature on discrete truss optimisation has an extensive history with many differing techniques being applied in a succession of increasingly more refined and iteratively better results. Automated truss design and improvement has been the subject of focussed resarch for over a century, stretching back to the pioneering work of A.G.M. Mitchell in 1904 [17]. Human performance has long been outstripped by iteratively better adapted algorithms. The results in papers (i) and (ii) are equal to or better than all examined techniques from the literature [14, 43, 44, 45, 46, 47, 48, 49] on all examined problems. ---------------------------------- G ("The result solves a problem of indisputable difficulty in its field.") The individual problems of truss topology optimisation (connectivity of members) and member sizing optimisation have long been considered difficult in isolation (topology optimisation is proven to be NP-Complete). The combination of both problems simultaneously (as shown in this paper) represents a problem of indisputable difficulty in the engineering field. Our approach is novel because we have changed the representation of the problem with respect to the classic structural topology optimisation literature. While classic ground structure discrete topology optimisation problems seek to find the optimal connectivity arrangement between fixed node locations, our approach mimics the far more computationally and structurally efficiant method of continuous topology optimisation, whereby superflous material is iteratively removed from a solid block, leaving a highly optimised, organic-looking structure. While such continuous structures are highly efficient, they are impractical and prohibitively expensive to manufacture. Thus, our approach mimics the ideology behind the continuous technique, but using standard discrete construction elements. The end result is a structure composed of simple discrete elements, but which is uhinhibited by the representational constraints of traditional discrete topology optimisation techniques. Furthermore, the use of a Delaunay triangulation algorithm for node connectivity ensures that all evolved structures are guaranteed to be kinematically stable. A number of notable differences are evident between the results from paper (ii) and the rest of the contemporary literature, which further increase the difficulty of the problem as solved by paper (ii): - the systems in both papers (i) and (ii) are constrained in their representational capacity by the fact that they can only select from a pre-defined discrete range of readily available construction elements rather than selecting the most purely optimal cross-section. This decision was made in order to more closely align the systems with existing construction practices, i.e. solutions are scope-limited in order to be capable of being constructed in the real world. - the systems in both papers (i) and (ii) are further constrained in their use of contemporary building standards and codes of practice for constraints on evolved solutions. These constraints decrease the overall number of "viable" solutions (i.e. solutions which pass all constraints) as the extra constraints imposed are tougher than those used by the literature. - all techniques polled from the contemporary literature used unrealistic physical assumptions about the form of the designed solutions, namely the fact that they allowed solid members to pass through one another with no physical effects. The system in paper (ii) was designed to be as physically accurate as possible, and as such is thus further constrained in its representational capabilities in comparison to other techniques from the literature. ---------------------------------- ---------------------------------- 7. a full citation of the paper (that is, author names; publication date; name of journal, conference, technical report, thesis, book, or book chapter; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable); @article{Fenton:2014:Truss, title={Automatic innovative truss design using grammatical evolution}, author={Fenton, Michael and McNally, Ciaran and Byrne, Jonathan and Hemberg, Erik and McDermott, James and O'Neill, Michael}, journal={Automation in Construction}, volume={39}, pages={59--69}, year={2014}, publisher={Elsevier} } @article{Fenton:2016:Truss, title={Discrete planar truss optimization by node position variation using grammatical evolution}, author={Fenton, Michael and McNally, Ciaran and Byrne, Jonathan and Hemberg, Erik and McDermott, James and O’Neill, Michael}, journal={IEEE Transactions on Evolutionary Computation}, volume={20}, number={4}, pages={577--589}, year={2016}, publisher={IEEE} } ---------------------------------- ---------------------------------- 8. a statement either that "any prize money, if any, is to be divided equally among the co-authors" OR a specific percentage breakdown as to how the prize money, if any, is to be divided among the co-authors; Any prize money, if any, is to be awarded to the lead author who will disseminate it amongst the co-authors. ---------------------------------- ---------------------------------- 9. a statement stating why the authors expect that their entry would be the "best"; There have been precious few paradigm shifts or large gains in performance in the overall field of truss optimisation. The field has long focussed on minute incremental improvements We feel that this entry would qualify as a winning entry for a number of reasons: - The work has been published in high-quality high-impact journals in both the field of evolutionary computataion (IEEE Transactions on Evolutionary Computation) and in the application field itself (Automation in Construction). - The methods described in paper (ii) represent an entirely new way to generate truss structures, previously unseen in the literature. The formulation of the system ensures that all generated solutions are kinematically stable, and as such are structurally viable (constraints notwithstanding). - The results from these papers represent the pinnacle of engineering optimisation in a field that is over 100 years old, producing results equal to or superior than those from the literature. - The problems attempted in this submission are significantly more difficult than those in the comparative literature, as real-world materials, constraints, physical limitations have been imposed. - Since the papers have focussed on real-world applications, the work has high commercial potential. - With environmental issues at the forefront of most fora, this work has the potential to make a significant impact and contribution to sustainable development. Minimizing material useage has a number of benefits: - Reduction in self-weight of a single structure reduces the stresses imposed on the rest of the structure, thus allowing the rest of the structure to be downsized, which in turn further reduces the weight of the overall structure. When multiple components of a structure can be optimised, the effect snowballs to produce large scale savings across the entire overall structure. - Minimising material usage has obvious cost benefits. However, what is not immediately obvious is the effect of supply and demand for the materials themselves. Minimising material usage diminishes demand, which in turn dictates a reduction in supply. With an average estimate of 2 tons of CO2 being emitted for every 1 ton of steel produced [Global CCS Institute, 2013], any reduction in production has significant environmental implications. ---------------------------------- ---------------------------------- 10. An indication of the general type of genetic or evolutionary computation used, such as GA (genetic algorithms), GP (genetic programming), ES (evolution strategies), EP (evolutionary programming), LCS (learning classifier systems), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc. GE (grammatical evolution), a grammar-based form of Genetic Programming. ---------------------------------- ---------------------------------- Sources Global CCS Institute, 2013. "CCS for iron and steel production". Available online at: https://www.globalccsinstitute.com/insights/authors/dennisvanpuyvelde/2013/08/23/ccs-iron-and-steel-production