Prof. Joong Hoon Kim
Harmony Search Variants and Their Application to Engineering Problems
Among optimization methods, metaheuristic algorithms have shown their capabilities for finding the near-optimal solution for which analytical methods may not be able to produce an acceptable solution within a reasonable computation time, especially when the global optimum is surrounded by many local optima. Therefore, the need of using these approaches is understood by the optimization community.
The Harmony Search (HS), published in 2001, is derived from the concepts of musical improvisations and harmony knowledge, and is considered as one of the well-known metaheuristic algorithms. So far, the HS has proved its advantages over other optimizers and there are many improved variations of the HS in the literature. In recent years, it has become obvious that the concentration on a sole optimization method is rather restrictive. A skilled combination of concepts from different optimizers can provide more efficient results and a higher flexibility when dealing with large-scale problems, especially with large scale engineering problems.
This keynote speech will first introduce the concept of Harmony Search and the potential abilities of its variants including improved and/or hybridized versions. In addition, applications to some typical engineering problems will be discussed along with the results.
Professor Kim, a faculty of Korea University in the School of Civil, Environmental and Architectural Engineering, obtained his Ph.D. degree from the University of Texas at Austin in 1992 with the thesis title “Optimal replacement/rehabilitation model for water distribution systems”. Prof. Kim’s major areas of interest include: optimal design and management of water distribution systems, application of optimization techniques to various engineering problems, and development and application of evolutionary algorithms.
His publication includes “A New Heuristic Optimization Algorithm: Harmony Search”, Simulation, February 2001, Vol. 76, pp 60-68, which has been cited over 3,300 times by other journals of diverse research areas. His keynote speeches include “Optimization Algorithms as Tools for Hydrological Science” in the Annual Meeting of Asia Oceania Geosciences Society held in Brisbane, Australia in June of 2013, “Recent Advances in Harmony Search Algorithm” in the 4 th Global Congress on Intelligent Systems (GCIS 2013) held in Hong Kong, China in December of 2013, and “Improving the convergence of Harmony Search Algorithm and its variants” in the 5 th International Conference on Soft Computing For Problem Solving (SocProS’ 2014) held in Saharanpur, India in December of 2014, “Performance of Evolutionary Algorithms on the Fitness Landscape of Specific Problems” in the 5th International Conference on Soft Computing For Problem Solving (SocProS’ 2015) held in Silcha, India in December of 2015, “Optimization Algorithms for Smart Water Management: which one to choose?” in the 12 th International Conference on Hydroinformatics (HIC 2016) in Incheon, South Korea in August of 2016. He hosted several conferences inclufing the 6 th Conference of The Asia Pacific Association of Hydrology and Water Resources (APHW 2013) and the 1 st and 2 nd International Conferences on Harmony Search Algorithm (ICHSA 2015).
Prof. Ajit Kumar Verma
A Computational Intelligence Framework for Optimal Maintenance of Large Engineering Systems
Maintenance Optimization for Large Engineering Systems is a key area in asset management. Machinery Health Monitoring is increasingly being adopted by industries not only as a means for asset management but also for ensuring high levels of vailability with its consequent gains by ascertaining the impact in enhancing availability at plant level and in a cost effective manner. Quantification of certain features such as detect-ability and prognostic ability and a fuzzy logic approach for doing so are presented. These features are then incorporated in a multi-objective maintenance optimization model based on Markov process and genetic algorithm, the results of which serve as decision support system for the selection of health monitoring. A demonstrative case is presented to illustrate this idea.
Ajit Kumar Verma is a Professor(Technical Safety) for about six years in Norway and is working with ATØM, Western Norway University of Applied Sciences, Haugesund, Norway. He was a Professor/Senior (HAG) scale Professor with the Reliability Engg Department of Electrical Engineering at IIT Bombay for around 15 years with a research focus in Reliability, Risk & Safety Engineering and Soft Computing Applications in various domains. He is also a Guest Professor at Lulea University of Technology, Sweden and was an adjunct at the University of Stavanger. He has been honored with ‘Honorary Professor’ at Amity University in India. He is the Springer Book Series Editor of Asset Analytics: Performance and Safety Management as well as the Springer Book Series Editor of Reliable & Sustainable Electric Power and Energy Systems Management and has jointly edited books published by Springer titled 1.Reliability and Risk Evaluation of Wind Integrated Power Systems 2.Reliability Modeling and Analysis of Smart Power Systems 3.Current Trends in Reliability, Availability, Maintainability and Safety- An Industry Perspective 4.Sustainable Power Systems and is also an author of books titled 1.Fuzzy Reliability Engineering Concepts and Applications (Narosa), 2.Optimal Maintenance of Large Engineering System-Practical Strategies for Effective Decision Making(Narosa), 3.Reliability and Safety Engineering (Springer) 4.Dependability of Networked Computer Based Systems (Springer), 5.Risk Management of Non-Renewable Energy Systems(Springer). He has over 250 publications in various journals and conferences and has been a supervisor for 37 Phd theses.He is a senior member of IEEE and a life fellow of IETE. He has been the Editor-in-Chief of OPSEARCH published by Springer(Jan 2008- Jan 2011) as well as the Founder Editor-in-Chief of International Journal of Systems Assurance Engineering and Management(IJSAEM) published by Springer and an Editor-in-Chief of Journal of Life Cycle Reliability and Safety Engineering(Springer). He is on the editorial board of various international journals. He has served as a Guest Editor of Special Issues of journals including IEEE Transactions on Reliability.
Dr. Veena Mendiratta
Anomaly Detection in Wireless Networks using Mobile Phone Data
Communications traffic on wireless networks generates large volumes of metadata on a continuous basis across the various servers involved in the communication session. Since these networks are engineered for high reliability, the data is predominantly normal with only a small proportion of the data being anomalous. It is, however, important to detect these anomalies when they occur because such anomalies are indicators of vulnerabilities in the network. In this talk we will present: non-parametric methods for anomaly detection, and the use of neural network based Kohonen Self Organizing Maps (SOM) and visual analytics for network anomaly detection and analysis using data from a 4G wireless network.
Veena Mendiratta is the Research Leader for Network Reliability and Analytics at Nokia Bell Labs based in Naperville, Illinois, USA. Her research interests include system and network dependability analysis, software reliability engineering, programmable networks (SDN) resiliency, and telecom data analytics. Current research is focused on network reliability and analytics – architecting and modeling the reliability of next generation programmable networks, and development of analytics-based anomaly detection algorithms for improving network performance and reliability. She is a member of the SIAM Visiting Lecturer Program, Life Member of SIAM, Senior Member of IEEE, Member of INFORMS; and was a Fulbright Specialist Scholar for 5 years during which time she visited universities in India, New Zealand, and Norway. She holds a B.Tech in engineering from IIT-Delhi, India, and a Ph.D. in operations research from Northwestern University, USA.
Dr. Seyedali Mirjalili
Confidence-based robust optimisation
Robust optimisation refers to the process of combining good performance with low sensitivity to possible perturbations. Due to the presence of different uncertainties when optimising real problems, failure to employ robust optimisation techniques may result in finding unreliable solutions. Robust optimisation techniques play key roles in finding reliable solutions when considering possible uncertainties during optimisation.
Evolutionary optimisation algorithms have become very popular for solving real problems in science and industry mainly due to simplicity, gradient-free mechanism, and flexibility. Such techniques have been employed widely as very reliable alternatives to mathematical optimisation approaches for tackling difficulties of real search spaces such as constraints, local optima, multiple objectives, and uncertainties. Despite the advances in considering the first three difficulties in the literature, there is significant room for further improvements in the area of robust optimisation, especially combined with multi-objective approaches.
In this talk, a brief discussion on preliminaries and key concepts of robust optimisation will be given first. The essential components of a systematic robust optimisation algorithm design process are then presented. Finally, the newly emerged confidence-based robust optimisation perspective will be discussed to find robust optimal solutions for engineering design problems with expensive objective functions. The talk will focus on both single and multi-objective robust optimisations. Also, a real application of the confidence-based robust multi-objective optimisation will be introduced as a practical demonstration.
Dr Seyedali Mirjalili is a lecturer in Griffith College, Griffith University. He received his B.Sc. degree in Computer Engineering(software) from Yazd University, M.Sc. degree in Computer Science from Universiti Teknologi Malaysia (UTM), and Ph.D. in Computer Science from Griffith University. He was an active member of Soft Computing Research Group (SCRG) at UTM and Institute for Integrated and Intelligent Systems (IIIS) at Griffith University. His research interests include Robust Optimisation, Engineering Optimisation, Multi-objective Optimisation, Swarm Intelligence, Evolutionary Algorithms, and Artificial Neural Networks. He is working on the application of multi-objective and robust meta-heuristic optimisation techniques in Computational Fluid Dynamic (CFD) problems as well.
Dr Mirjalili is internationally recognised for his advances in Swarm Intelligence (SI) and optimisation, including the first set of SI techniques from a synthetic intelligence standpoint – a radical departure from how natural systems are typically understood – and a systematic design framework to reliably benchmark, evaluate, and propose computationally cheap robust optimisation algorithms. Dr Mirjalili has published over 50 journal articles, many in high-impact journals, with one paper having over 500 citations – the most cited paper in the Elsevier Advances in Engineering Software journal. In addition he has more than a dozen book chapters and conference papers. Dr Mirjalili has over 2000 citations in total with an H-index of 21 and G-index of 45. From Google Scholar metrics, he is globally the 6th most cited researcher in Engineering Optimisation and the 9th most cited in Robust Optimisation.
His research output speaks for itself – Dr Mirjalili is clearly an emerging leader in his field. He has received numerous awards for his work, including two from the Institute for Integrated and Intelligent Systems for “Most Significant Research Impact” and “Best Publication in Artificial Intelligence”. He was awarded a highly competitive international postgraduate scholarship and was fully supported by Griffith University. With 30 submissions and an average of over 1000 downloads per month, Dr Mirjalili has been an active member in Matlab community and a top-30 Matlab File Exchange contributor worldwide since 2013. He was awarded Matlab central challenge coin as an outstanding contributor in 2015 and 2016 for the significant impact of his research outputs and dedication to research in the field.
In addition to his excellent research outputs, Dr Mirjalili also has an emerging profile as a teacher and mentor. He has worked as a course coordinator at Griffith for over five years and has taught in four subjects across two different program areas. He received an award for “Star Rising Teacher” from Griffith in 2015 for improving the quality of teaching and learning significantly in his subjects.
Prof. Ramakrishnan Ramanathan
Barriers and facilitators to Big Data and Analytics with an example study of using statistics in logistics and e-commerce
This invited talk will be based on a series of research projects to understand the diffusion and impact of Big Data and Analytics (BDA) in UK retail and logistics firms. Several insights on the soft aspects of implementing BDA, including barriers and facilitators, will be discussed. Finally, in line with the theme of the RTORS conference, an example application of statistics for collection and analytics in the context of logistics in e-commerce will be presented.
Professor Ram Ramanathan is the Director of Business and Management Research Institute, in the Business School of the University of Bedfordshire, Luton, UK. In the past, he has worked and taught in a number of countries, including the UK, Finland, the Netherlands, Oman and India. He has taught basic and advanced courses on Operations Management, Production Systems Management, Supply Chain Management, Optimization Theory, Data Envelopment Analysis (DEA), Management Science, Business Statistics, Simulation, Energy and Environment, Energy and Environmental Economics, Energy and Transport Economics, and others. His research interests include operations management, supply chains, environmental sustainability, economic and policy analysis of issues in the energy, environment, transport and other infrastructure sectors. He works extensively on modelling using techniques such as optimisation, decision analysis, data envelopment analysis and the analytic hierarchy process.
Ram has successfully completed a number of research projects across the world. He is on the editorial boards of several journals and in the technical/advisory committees of several international conferences in his field. He is an advisory board member of an innovative new online resource, The Oxford Research Encyclopedia of Business and Management. He is a member of ESRC Peer Review College in the UK and reviews applications submitted to the Newton Fund. He has produced five books (including a new book on Big Data Analytics using Multi-Criteria Decision Making, and an introductory textbook on DEA), more than 126 research publications in journals and more than 156 conference presentations. His research articles have appeared in many prestigious internationally refereed journals includingOmega, Journal of Business Ethics, Tourism Economics, International Journal of Production Economics,Supply Chain Management,International Journal of Operations & Production Management, European Journal of Operational Research,Transport Policy, and,Transportation Research.
Dr. Atma Sahu
Soft Computing Methods and Teaching: Pathway to Research Collaborations
Computational methods and mathematical modeling is fundamental and essential in solving advance engineering and natural sciences problems in high priority areas where multidisciplinary and multi-institutional expertise is required. This presentation explore a research agenda with position of one or more areas of computational methods applications in elastic vibrations of beams and artificial intelligence (machine learning). Additionally, participants interested in genetic algorithms focused to generate high-quality solutions to optimization and search problems by relying on bio-inspired operators are in the research plans. Lastly, the presentation will conclude with a brief direction and pathway in designing and collaborating multidisciplinary and multi-institutional research fund opportunities with NSF, IUSSTF, DST, SERB, CSIR etc., which are just a few national and international agencies sources, and explore suitable grants opportunities for start-up research and for using the scientific and industry expertise in collaborative team format. Interested beginner or veteran proposals writing investigators from interdisciplinary areas are encouraged to develop collaborative research teams. These teams plan critical mathematical and computational techniques using math- modeling and algorithm development to be able to make known accurate, efficient and reliable solutions through implementation.
Dr. Atma Sahu currently serving as a Professor in Mathematics, Coppin State University , Baltimore MD USA. His research work areas include vibrations of elastic beams, computational mathematics and mathematical modeling such as genetic algorithms and machine learning. Additionally, he is well-recognized researcher in educational aspects systems approach, curriculum and instruction design in STEM fields, and related assessment pedagogy. He also consult with Dept of Defense, Dept of Education, University System Of Maryland institutions of higher Ed. Dr. Sahu has published numerous research papers in US domestic and international refereed journals. He is a member of IEEE USA, AMS and NCTM.
Dr. R. N. Mohapatra
Correlation Pattern Recognition
Filters are very useful in signal analysis. We have seen high pass and low pass filters and wavelet filter banks etc. . Correlation filter theory relies heavily on concepts and tools from the fields of linear algebra and probability theory. Matrices and vectors provide succinct methods of expressing operations on discrete images, manipulating multiple variables and optimizing multiple parameters. In this talk we will introduce the notion of correlation filters and quadratic filters for pattern recognition, face recognition and its other applications. It is an interesting topic where mathematicians, statisticians and engineers can find avenues for new research. We will demonstrate the results of a case study associated with automatic pattern recognition.
Dr. R. N. Mohapatra graduated with a Ph.D. in Mathematics in 1968 from Jabalpur University under the guidance of late Professor Tribikram Pati. He worked as a lecturer at the Regional College of Education, Bhubaneswar and the Post Graduate Department of Mathematics, Sambalpur University before joining the American University of Beirut, Lebanon. He visited the University of Alberta, Alberta, Canada, where he worked with Dr. Ambikeswar Sharma on Spline Approximation. In 1983 he went to York University, Ontario, Canada to teach Statistics.
From August 1984 he had been working at the University of Central Florida, where he is a Professor of Mathematics. He has published over 140 research papers in peer reviewed journals in Summability Theory and sequence spaces, Fourier Analysis, Approximation and Geometric Function Theory, Fluid Mathematics, Mathematical Physics, Mathematical models in Epidemiology, Frame Theory in Hilbert C* modules, Variational Inequalities and Optimization Theory. He is has written two books and seven edited volumes. His current interest is to study “Image Encryption and Cyber Security”.