Decision and Cognitive Sciences Research Centre

News and events

News and events

DCS Seminar Series

 

Date: Wednesday, 26 September 2018

Time: 1pm - 2pm

Location: B4 AMBS East

Title: Big data for health management - Early Diagnosis, Intervention & Prevention

Presenter: Dr Xin Shi

Abstract:

Healthcare is the most important demand of humans. Long and healthy life is one of the primary research subjects in human health research. However, it is difficult to accurately assess health status at a very early stage, with the aim of determining appropriate interventions to maintain good health and wellbeing. Therefore, it is essential to optimise human health management polices and assess the risk factors associated with health status. Human health management is the process and means for health risk factors monitoring, prognostics, intervention and control based on our knowledge on human health and disease prevention using non-clinical and clinical linkage data. Some symptoms that could indicate potential advanced disease or chronic disease may often be ignored or missed. This will lead to serious delay in clinical diagnosis and treatment intervention. Subsequently, it will increase the medical treatment costs as well as patients physical, mental and financial burden. Our study aims to develop a systematic approach which integrates statistical modelling of health big data into optimal health management decision-making with mobile application. By developing statistical modelling method for health big data on early diagnosis, prevention and intervention, we are developing a multi stage delay-time model to investigate risk factors and predict heath status at an early stage of disease/illness progression using linked clinical and non-clinical data. In this talk, we will present our recent research outcomes and discuss the challenges for the future study.

Short Bio

Dr Xin Shi was trained in statistics at Salford University of. He also has post-doctoral researcher and lecturer experiences at the school of medicine of both Sheffield University and Manchester University. Dr Xin Shi is currently Reader at Manchester Metropolitan University and leads a team in collaborating with China. Xin is qualified Chartered Statistician awarded by the Royal Statistical Society in the UK. His main research interest lies in the field of business analysis for life course modelling in healthcare services. His current active research focuses on using non-clinical and linked big data to predict clinical outcomes on personalised health management and for decision-making, e.g. cancer early diagnosis. His current research projects cover: a) cancer early diagnosis using non-clinical big data; b) Chinese wellbeing/smoking big data with survival analysis; 3) Brazilian well-being big data study; d) delay-time model on human health management, in particular for chronic disease; e) British drug related death study. Xin is experienced on health related big data including linked data in the UK, China and Brazil, in particular for the relevant statistical methods. Xin has published in journals including Scientific Reports, European Journal of Operational Research, Statistical Methods for Medical Research, Journal of Royal Statistical Society A, Otology and Neurology, etc. One of his papers published in Cochrane Database of Systematic Reviews has been cited over 300 times. In the last five years, he has won about £900k research grants from research councils, Newton Fund and the Innovation UK (for Knowledge Exchange Projects) Principle Investigator or Co-Investigator. Dr Xin Shi is a managing or associate editor of a number of top journals. Xin has also awarded fellowship of Royal Statistical Society and High Education Agency. He also serves on grant review committee for research funding bodies, such as British Council, NIHR, UNESCO.

Date: Monday, 17 September 2018

Time: 10:00 – 11:00

Location: B5 AMBS East

Title: Research Sampling at the Networking and Emerging Optimization Group (NEO)

Presenter: Prof. Enrique Alba, University of Málaga (Spain)

Abstract:

This talk will develop on the many different topics that the group NEO (http://neo.lcc.uma.es) has been dealing with along the recent years. We will present results in research, development, and innovation, based in extending the state of the art in metaheuristics, machine learning, and advanced algorithms in general.

The topics of application will include the domains of telecoms, smart mobility, bioinformatics, and software engineering, to name a few. To face these scientific and real world problems we have been defining new solvers based in HPC (clusters, grids, multicores, GPUs), dynamic optimization, multi-objective advances, and theoretical descriptions of problem landscapes. All together will bring a wide survey of challenging academic problems, real applications / services, and new techniques in search, optimization, and learning.

Short Bio

Prof. Enrique Alba had his degree in engineering and PhD in Computer Science in 1992 and 1999, respectively, by the University of Málaga (Spain). He works as a Full Professor in this university with different teaching duties: data communications, distributed programing, software quality, and also evolutionary algorithms, bases for R+D+i and smart cities at graduate and master/doctoral programs. Prof. Alba leads an international team of researchers in the field of complex optimization/learning with applications in smart cities, bioinformatics, software engineering, telecoms, and others. In addition to the organization of international events (ACM GECCO, IEEE IPDPS-NIDISC, IEEE MSWiM, IEEE DS-RT, …), Prof. Alba has offered dozens postgraduate courses, multiple seminars in more than 30 international institutions, and has directed several research projects (7 with national funds, 5 in Europe, and numerous bilateral actions). Also, Prof. Alba has directed 7 projects for innovation in companies (OPTIMI, Tartessos, ACERINOX, ARELANCE, TUO, INDRA, AOP) and presently he also works as invited professor at INRIA, the Univ. of Luxembourg, and Univ. of Ostrava. He is editor in several international journals and book series of Springer-Verlag and Wiley, as well as he often reviews articles for more than 30 impact journals. He has published 90 articles in journals indexed by Thomson ISI, 17 articles in other journals, 40 papers in LNCS, and more than 250 refereed conferences. Besides that, Prof. Alba has published 11 books, 39 book chapters, and has merited 6 awards to his professional activities. Prof. Alba’s H index is 46, with more than 10,000 cites to his work.


Date: Wednesday, 12 September 2018

Time: 13:00 - 14:00

Location: B5 AMBS East

Title: Deep Reinforcement Learning: Frontiers, Challenges and Applications

Presenter: Dr. Arjun Chandra

Abstract:

Reinforcement Learning (RL) is a conceptual framework allowing the study and synthesis of autonomous agents that can learn and plan to make far-sighted decisions from their experiences. These experiences are a result of agents attempting to solve problems interactively, thereby learning from feedback on their solutions, followed by adapting these solutions. The RL framework primarily forges solutions to control problems. These solutions take the form of agent behaviours -- action/decision sequences or plans. Crucially, the control problems are such that actions executed by an agent have delayed consequences. Furthermore, the dynamics of such problems are hard to model analytically. A robot that can walk, a car that can drive itself, a drone that can fly itself, a gaming agent that scores high/wins, an industrial plant that controls resource use to reduce long term energy consumption, a customer relationship agent that interacts with customers to keep them satisfied, etc., are some examples fitting the framework. Problems in the telecommunications, health, and education sectors, amongst others, also fit the framework. Up until recently, theoretically sound RL algorithms were hard to apply in practice. One reason for this was the sheer number of problem states, e.g. observations via sensor measurements, which explode combinatorially as the problem becomes more realistic. This needed additional considerations. Deep learning and innovations in training regimens came to the rescue. Thus began the rise of what is now referred to as deep reinforcement learning (deep RL). This seminar will introduce deep RL and attempt to draw a comprehensive picture of the field as it stands. Work from various academic and industrial research labs will be covered from an intuitive point of view. In addition, the historical development of the field will be examined, building up to current frontiers. The seminar will also aim at evoking a sense of appreciation for the fundamental research challenges in the field.

Short Bio

Dr. Arjun Chandra received a PhD in Computer Science from the University of Birmingham in 2011. He currently works as a Senior Research Scientist at Telenor Research, the R&D wing of a Norwegian multi-national telecommunications company headquartered in Oslo. His research interests broadly lie within the domains of machine learning and multi-agent systems. In recent years, his research has focussed on engineering game theoretic and psychology inspired approaches to self-organisation in multi-agent systems, and on investigating fundamental challenges with scaling reinforcement learning. Arjun is also affiliated with the Norwegian Open AI Lab, through which he routinely advises graduate students. Arjun has previously worked as a Research Associate/Post-doc at the Universities of Manchester, Birmingham, and Oslo, on projects funded by the UK research council (EPSRC) and the EU. Before joining Telenor, he also helped build the automated tutoring technology at Studix, an education technology company.

Date: Wednesday, 18th July 2018

Time: 13:00 - 14:00

Location: B4 AMBS East

Title: Challenges for Interactive Multiobjective Optimization in Natural Resources Planning – Some Numerical Experiments

Presenter: Emeritus Prof Theodor Stewart

Abstract:

Strategic natural resources planning typically involves multiple (“many”) objectives that need comparative evaluation by groups, often involving subjective considerations. We describe the role of the algorithm as generating a potential shortlist (“7 +/- 2”) of alternatives for such evaluation. We motivate a useful role for simultaneous reference point methods, firstly to generate a dispersed set of Pareto optimal solutions. Then, after a (possibly only partial) selection is made from these, a few more alternatives may need to be generated in an interactive manner, taking into account the preferences revealed from earlier sets of comparisons, by refining the set of reference points.

Within the context of reference point methods, we discuss experiments to evaluate in particular (a) design of a set of reference points to represent an initially wide spread of potential preferences; and (b) progressive refinement of this design in the light of preferences expressed (after which the process repeats).

Short Bio

Theodor Stewart is Emeritus Professor of Statistical Sciences at the University of Cape Town South Africa, and continues to hold a part time post as senior research scholar at that institution.  From 2007-2017 he was part-time professor of decision science at AMBS.

His book with Valerie Belton onn Multiple Criteria Decision Analysis has (per Google Scholar) attracted over 3500 citations.  He is editor-in-chief of the Journal of Mult--Criteria Decision Analysis and an associate editor of internatioal transactions in operational research.

Date: Friday, June 22, 2018

Time: 13:00 - 14:00

Location: B7 AMBS East
 
Title: Thermal conductivity enhancement of thermal energy storage

Presenter: Dr Heinrich Badenhorst

Abstract:

Thermal energy storage (TES) systems are used to accumulate energy for use in both heating and cooling applications. Interest in TES is growing rapidly for a variety of reasons, including the incorporation of highly variable, renewable resources into the energy grid. These systems are mainly used to mitigate differences between the supply of and demand for, energy. In this capacity they undergo three operational steps: charge, store and discharge. The speed and frequency at which the cycle is operated depends on the application. For example, at a solar power plant it may be beneficial to have a storage unit that rapidly discharges to compensate for momentary gaps in sunshine due to cloud cover. On the other hand, if the objective is to smooth the effect of peak consumer usage in the late afternoon and evening, a store which discharges at constant rate, over the time frame of several hours might be desired. Recent work has been focused on using computational fluid dynamics as an aid for designing and optimising transient thermal systems. From a design perspective it is essential that a methodology is used which considers the fact that these stores operate more like thermal batteries than heat exchangers. This requires a departure from conventional steady-state approaches to a technique that directly incorporates the required dynamic behaviour of the system. Machine learning, in the form of genetic algorithms, has been combined with finite volume based simulation of simplified, sensible energy stores. As preliminary trial, the approach has been applied to the design of free-form, non-parametric heat exchanger fins to optimise the thermal output of a rectangular store. It is hoped that this approach can be extended to phase change based storage and multi-component composites.

Short Bio

Dr Heinrich Badenhorst is a lecturer in the School of Chemical Engineering and Analytical Science at the University of Manchester. His industrial experience is in process modelling, optimization and control, acquired at Sasol Synthetic Fuels in South Africa and numerous U.K. and European petrochemical plants. Following his work in industry he completed his PhD on modelling the fundamental mechanisms which govern oxidation in nuclear grade graphite. His current research interests’ lie in the development of carbon materials for use in energy based applications, specifically thermal energy storage, solar capture, desalination, thermo-chemical energy storage and next generation nuclear reactors. Special focus is on the development of high thermal conductivity additives and the use of CFD simulations for latent heat based energy storage design. Recent work is aimed at using reversible metal hydroxide-oxide reactions to capture and store energy.

Date:  Friday, June 15, 2018

Time: 13:00 - 14:00

Location: B4 AMBS East

Title: The application of cooperative games in priority queues and kidney exchange

Presenter: Mr. Liu Hanlin

Abstract:

We study resource sharing whereby multiple independent service providers collaborate by pooling their resources into a single service facility and deploying an optimal scheduling policy for their customers collectively by taking account of their individual waiting costs and service level requirements. We model the pooled systems as M/M/c queueing systems with heterogeneous waiting costs subject to service level constraints. We investigate cost allocation rules to the pooled systems by applying concepts from cooperative game theory. We consider both the case with a fixed number of servers and the case with endogenous capacity optimization. To empower our analysis, we show that a cooperative game with polymatroid optimization can be analyzed through simple auxiliary games. By exploiting analytical properties of the continuous extension of the total virtual workload and multiclass queueing systems with polymatroidal structures, we provide sufficient conditions for the games to possess a core allocation. If the cores of the cooperative games do not exist, we also provide approximate stable allocations. Finally, we explore the extent that our results remain valid for general multiclass queueing systems. I will also talk something about the application of cooperative game theory in the kidney exchange problem.

Short Bio

Mr. Liug Hanlin is a final year PhD candidate from the department of System Engineering and Engineering Management, City University of Hong Kong. Before that, he get the B.S degree in Applied Mathematics from the University of Science and Technology of China. His research interests includes cooperative game theory, system reliability modelling and inventory management. 

Date: Friday, 1st June 2018

Time: 12:30 - 13:30

Location: B5 AMBS East

Title: Exact and Hybrid Approaches for Packing While Traveling and the Traveling Thief Problem

Presenter: Prof Frank Neumann

Abstract:

Multi-component problems play a crucial role in real-world applications, especially in the area of supply chain management. Recently, the traveling thief problem (TTP) has been introduced to study multi-component problems in a systematic way and many heuristic search algorithms have been proposed for the TTP. Although a lot of algorithmic advances have been made on this problem, determining an optimal solution, even for small instances, is very challenging. In this talk, we will present exact and hybrid approaches for this problem. We start by investigating the already NP-hard Packing While Traveling (PWT) problem which results from TTP when the TSP tour is fixed. We present an exact dynamic programming approach for PWT and give a fully polynomial time approximation scheme (FPTAS) for PWT over its baseline travel cost. Afterwards, we extend the approach to give a dynamic programming (DP) approach for TTP and report on some experimental results. Furthermore, we will show how the DP for PWT can be incorporated into an evolutionary multi-objective algorithm to tackle a multi-objective formulation of TTP. Joint work with Sergey Polyakovskiy, Martin Skutella, Leen Stougie, Junhua Wu, Markus Wagner.

Short Bio

Frank Neumann is a professor and leader of the Optimisation and Logistics Group at the School of Computer Science, The University of Adelaide, Australia. He received his diploma and Ph.D. from the Christian-Albrechts-University of Kiel in 2002 and 2006, respectively.  Frank has been the general chair of the ACM GECCO 2016. With Kenneth De Jong he organised ACM FOGA 2013 in Adelaide and together with Carsten Witt he has written the textbook "Bioinspired Computation in Combinatorial Optimization - Algorithms and Their Computational Complexity" published by Springer. He is an Associate Editor of the journals "Evolutionary Computation" (MIT Press) [ABS 3] and "IEEE Transactions on Evolutionary Computation" (IEEE) [ABS 4]. In his work, he considers algorithmic approaches in particular for combinatorial and multi-objective optimization problems and focuses on theoretical aspects of evolutionary computation as well as high impact applications in the areas of renewable energy, logistics, and mining.

Date: Wednesday, 23rd May 2018

Time: 13.00 - 14.00

Location: B4 AMBS East

Title: Game-Theoretical Fair Decision-Making using Mathematical Programming

Presenter: Dr Songsong Liu

Abstract:

Fair decision-making has been widely investigated in various fields, e.g., economics, engineering and social science, in which concept, perception and interpretation of fairness vary depending on the problems dealt with. Meanwhile, mathematical programming is a powerful decision-making tool that has been applied to a great variety of theoretical and applied problems in the areas of engineering, energy, environment, business, finance, and management, etc. This talk will address the modelling and solution of game-theoretical fair decision-making using mathematical programming techniques, especially mixed integer programming techniques. Two fair decision-making approaches based on Nash bargaining and lexicographic maximin principles, under widely accepted proportional and max-min fairness criteria, respectively, are introduced for cooperative game and multiobjective optimisation problems. Case studies of supply chain planning and energy management of smart homes are investigated as the applications of the proposed effective and efficient mixed integer linear/nonlinear programming models and solution approaches.

Short Bio

Dr Songsong Liu is a Lecturer in Management & Systems in the School of Management, Swansea University. He received BSc degree in Information and Computing Science and MSc degree in Mathematics from Tsinghua University, China, and PhD degree in Chemical Engineering from UCL. He has worked as a Research Associate at UCL before joining Swansea University in 2016.  His research interests are in the development of optimisation-based decision-making models, algorithms and approaches for real applications in a range of areas, including supply chain management, production planning and scheduling, sustainable systems engineering, process design and synthesis, data mining and analytics. He has published about 50 peer-reviewed publications in these areas in international journals, books and conferences. He has been awarded with IChemE Hutchison Medal, RSC Emerging Talent Prize, UK Government Overseas Research Students Award, and Chinese Government Award for Outstanding Self-Financed Students Abroad, etc.

Date: Monday, 16th April 2018

Time: 11:00 - 12:00

Location: B5 AMBS East

Title: Predictive Analytics & A.I. for Good (Intelligent Disease Surveillance & Control)

Presenter: Prof Jiming Liu

Abstract:

In this talk, I will present some of the recent developments in the field of computational healthcare. Specifically, I will discuss the increasingly important role of data analytics and modeling in achieving epidemiological intelligence for controlling and preventing infectious diseases, one of the major challenges in today’s global health. I will identify several complex systems research issues as well as computational methods for characterizing and inferring disease risks that often involve multiple interacting factors at various spatio-temporal scales.

Short Bio

Jiming Liu is the Chair Professor of Computer Science and Associate Vice-President (Research) at Hong Kong Baptist University.  He received his M.Eng. and Ph.D. degrees in Robotics from McGill University, having obtained Master of Arts from Concordia University and B.Sc. from East China Normal University.  His current research interests include: Complex Systems and Autonomy-Oriented Computing; Complex Networks and Web Intelligence; and Computational Healthcare and Sustainability.  He is a Fellow of the IEEE, and was the Chair of IEEE Computer Society Technical Committee on Intelligent Informatics.  He has served as Editor-in-Chief of Web Intelligence Journal (IOS), and Associate Editor of IEEE Transactions on Knowledge and Data Engineering, IEEE Transactions on Cybernetics, Big Data and Information Analytics (AIMS), Neuroscience and Biomedical Engineering (Bentham), and Computational Intelligence (Wiley), among others.

Date:  Wednesday, 7st March 2018

Time: 13:00 – 14:00

Location: B4 AMBS East

Title: Applying Advanced Analytical Methods in Business – A Personal Perspective

Presenter: Dr Max Griffiths

Abstract:

The presentation investigates the use of more advanced analytics, such as predictive modelling and optimisation, within Financial Services and Utilities. Drawn from personal experience across an analytical career in these sectors, I'll explore what's important in an analysis, the benefit of making mistakes, understanding when to be accurate (and when not to be!) and the role of leadership and the team. The presentation includes a number of examples of good and bad analysis, developments and projects to illustrate the key points.

Short Bio

Dr Max Griffiths has more than 20 years' experience of leading analytical teams to deliver customer focused, credit risk and collections strategies. He began his career investigating public sector effectiveness with the Home Office, West Midlands Police and Fire Service, but has predominantly worked in the private sector for companies such as HSBC, Merrill Lynch and Centrica. He now works for a Specialist Mortgage lending, Together, where he is Head of Credit Risk & Analytics. Data-driven surrogate-assisted evolutionary optimization of expensive optimization problems.

Date:  Wednesday, 31st January 2018 Time: 13:00 – 14:00

Location:: B2 AMBS East

Title: Data-driven surrogate-assisted evolutionary optimization of expensive optimization problems

Presenter: Prof. Yaochu Jin

Abstract:

This talk discusses the main challenges in data-driven surrogate-assisted evolutionary optimization of expensive problems. Fundamental issues such as surrogate model selection, surrogate model management and model training using advanced machine learning techniques in single and multi- and many-objective optimization will be discussed. Challenges and recent advances in surrogate-assisted optimization of high-dimensional expensive optimization problems will be presented. Finally, a few real-world examples will be given.

Short Bio:

Prof. Yaochu Jin received the B.Sc., M.Sc., and Ph.D. degrees from Zhejiang University, Hangzhou, China, in 1988, 1991, and 1996 respectively, and the Dr.-Ing. degree from Ruhr University Bochum, Germany, in 2001. He is a Professor in Computational Intelligence, Department of Computer Science, University of Surrey, Guildford, U.K., where he heads the Nature Inspired Computing and Engineering Group. He is also a Finland Distinguished Professor and a Changjiang Scholar. His main research interests include evolutionary computation, machine learning, computational neuroscience, and evolutionary developmental systems. He has published over 200 journal and conference papers and has been granted eight US/EU/Japan patents.Prof Jin is the Editor-in-Chief of the IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, and Co-Editor-in-Chief of Complex & Intelligent Systems. He is an IEEE Distinguished Lecturer (2017-2019, 2013-2015) and was the Vice President for Technical Activities of the IEEE Computational Intelligence Society (2014-2015). He is the recipient of the 2018 “IEEE Transactions on Evolutionary Computation Outstanding Paper Award”, the 2014 and 2016 “IEEE Computational Intelligence Magazine Outstanding Paper Award”, the 2010 IEEE CIBCB “Best Paper Award”, the 2007 IEEE FOCI, and the 2017 IEEE CEC “Best Student Paper Award”. He is a Fellow of IEEE. Data-driven surrogate-assisted evolutionary optimization of expensive optimization problems

Date:  Wednesday, 24th January 2018

Time: 13:00 – 14:00

Location: B4 AMBS East

Title: Longest-Queue: the long and short of it

Presenter: Dr. Neil Walton

Abstract:

When arriving at a set of queues, it is natural to want to join the shortest and, when serving queues, it is natural to want to serve the longest. We separately study Joint the Shortest Queue and Longest Queue First service. We discuss the consequences and generalizations of these decision rules. We discuss recent applications, motivating technologies and non-standard asymptotics. We apply contemporary methods to understand optimal and pessimal performance in these systems.

Short Bio:

Neil Walton studied his undergraduate ('05), masters ('06) and PhD ('10) in mathematics at the University of Cambridge. Following this he was a lecturer in mathematics at the University of Amsterdam. Since January 2016, he has been working as a senior lecturer at the School of Mathematics at the University of Manchester. In addition he has visited as a researcher at Microsoft Research Cambridge, the Basque Centre for Applied Mathematics and the Automatic Control Laboratory, ETH Zurich. He has twice been awarded best paper awards at the ACM Sigmetrics conference. He sat on council of the INFORMS Applied Probability Society and is an associate editor at Operations Research Letters.

Date:  Wednesday, 22nd November 2017

Time: 13:00 – 14:00 with light lunch provided from 12:30   

Location: B4 AMBS East

Title: Modelling of Tailored Nudging and Using It for Soft Influence in Information Security

Presenter: Dr. Iryna Yevseyeva

Abstract:

For some time behavioural change has been a focus of research in healthcare, e.g. for interventions to increase physical activity and healthy eating, to support smoking cessation and moderate drinking. Influencing behaviour in a paternalistic way has also been reported as efficient in other areas, such as tax payment and social policy design. Richard H. Thaler, the recent Nobel Prize winner in economics, persuaded many economists to pay more attention to human behaviour. In his works, e.g. famous “Nudge” book written together with C. Sunstein, he emphasised irrationality of human choices and easy ways to influence those. He showed that subtle changes, e.g. in the presentation of alternatives makes choosing some of alternatives easier to a decision maker. In the present work, an attempt to model influence using multicriteria decision aiding/analysis tools is presented. In particular, several design nudges were taken into consideration when developing multicriteria model and then tailored influence was developed for decision makers with different preferences. The model is applied to derive optimal Bring Your Own Devices security policies for the selection of Wi-Fi networks for employees using a device in a café, hotel or other public place, potentially less secure when compared to companies local networks. A company has the right to protect its employees from potential spoofing and influence their choices in a paternalistic way. It is demonstrated how the graphical user interface of the device may present choices so as to highlight most secure ones and also tailor influence towards those choices based on preferences elicited from previous choices of the user.

Date:  Wednesday, 15th November 2017

Time: 13:00 – 14:00

Location: E20 AMBS East

Title: The Legal-Economic Analysis of Computer Crime

Presenter: Dr. Len Noriega

Abstract:

Legal-Economic Analysis (LEA) is an area of research which combines elements of decision theory, data science and law in order to determine better policies at the national level. At a more mundane level the assumption of individual rationality can serve to improve policy measures within individual organisations. The talk will give an overview of the broader area of LEA and its application to computer crime, before going on to discuss approaches to safeguarding accountability and security within organisational units.

Short Bio

Dr. Len Noriega (BA MSc PhD LLB LLM) is currently MD of Data Burrowing Solutions Ltd in Stafford (UK), and Company Secretary of Al Hikmah Educational Consultancy, Umm Al Quwain, UAE. A history graduate from UCL in 1989, Len entered the field of computer science, graduating with an MSc from the University of Kent in 1990. Having worked in Engineering Computing Len moved into Artificial Intelligence studying Machine Perception at Keele University in 1993, and thence to complete his doctorate, working at the Polytechnic Universities of Valencia and Cartagena and Nottingham Trent University (UK). He completed post docs in colour physics at Keele, UMIST and Derby Universities. Worked as a university lecturer at Staffordshire University (2002-2013). He began teaching law to Computer Science students having studied the subject himself on a part time basis. In the Summer of 2013 he moved into a management role at a vocational college in Abu Dhabi (UAE) in 2013. In January 2017 he helped set up the Al Hikmah Educational Consultancy, and in August 2017 took over Data Burrowing Solutions Ltd.

Date:  Wednesday, 8th November 2017

Time: 13:30 – 14:30

Location: B7 AMBS East

Title: O.R. and the OR Society: Opportunities and Challenges

Presenter: Ruth Kaufman, President of the OR Society

Abstract:

The OR Society, with around 12 FTE staff and a turnover around £1million, falls definitely into the ‘S’ subset of SMEs. With 3,000 members, it is pretty small as far as learned societies go, as well. But its reach, impact and ambitions are disproportionately great. This discrepancy mirrors the nature of O.R. itself: a field defined by a huge aspiration (to address any and all problems of management), and a sweeping approach (the scientific method), even whilst most people who benefit from it have never heard of it, and many people who work in the field hesitate to define it.

In this talk, I will review some aspects of the current position of the ORS and of OR in the UK and discuss how professionals in O.R. and overlapping or neighbouring disciplines can best capitalise on our strengths and opportunities.

Short Bio

Ruth Kaufman OBE, FORS, FIMA is a Companion of OR and currently President of the OR Society. She graduated with a BA in Maths and Sociology from the University of Sussex in 1974, and worked in O.R. and other management functions in London Transport and London Electricity before joining the Department of Health as a Principal OR Analyst. She moved on to lead the OR Group at the UK Export Credits Guarantee Department (ECGD), and was Chair of the Government OR Service for two years. Before retiring from government in 2008, she was a member of ECGD’s Executive Board, as Head of Strategy, Change and OR.

Ruth was a co-founder of the ‘Making an Impact’ practitioners’ streams at the annual OR conference (and now at EURO conferences), and of the OR Pro Bono initiative where she continues to play a leading role. She was awarded an OBE in the 2016 New Years' Honours, for services to O.R.

As well as leading the OR Society, Ruth is currently a Visiting Senior Fellow at the London School of Economics, Chair of the EURO Working Group on Practice of OR, and undertakes a variety of other activities including freelance consulting.

Date:  Wednesday, 1st November 2017

Time: 13:30 – 14:30

Location: B4 AMBS East

Title: Consensus Building in Large-Group Decision-Making under Uncertainty: Challenges on Using Intelligent Decision Aid Approaches

Presenter: Iván Palomares Carrascosa

Abstract:

Decision-making is an inherent mankind process of ubiquitous nature in our daily lives. Real-life decision situations typically involve added complexities such as: (I) the need for effectively handling uncertainty stemming from human vagueness and subjectivity in expressing preferences; (II) the presence of multiple evaluation criteria and participants with diverse background, demanding appropriate preference aggregation methods; and importantly, (III) the importance of making highly accepted collective decisions in collective settings. All the above challenges accentuate in problems involving large, highly heterogeneous groups of decision makers. Large-group decision situations - in which participants may not necessarily be physically congregated- have increasingly become a reality in recent years, due to the rise of social network platforms and advances in mobile/cloud computing.

This talk firstly introduces some noteworthy challenges and limitations to support consensus building in decision-making problems involving large groups, followed by a categorisation of existing solutions based on intelligent techniques. Secondly, recent works undertaken by the speaker are presented on monitoring the preferences and behaviour of decision makers, and using agent-based techniques to aid decision makers in adjusting their preferences. Thirdly, ongoing work is presented in detail, namely on exploring subgroup clustering approaches and the influence of cognitive/behavioural aspects such as reliability and preference manipulation. The talk concludes with a series of “lessons learnt” and future directions of research.

Short Bio

Iván Palomares Carrascosa is a Lecturer in Computer Science with the School of Computer Science, Electrical and Electronic Engineering, and Engineering Maths (SCEEM), University of Bristol. He received his MSc and PhD degrees (with nationwide distinctions) from the Universities of Granada and Jaén (Spain). He has also been postdoctoral researcher and senior engineer with Ulster University and Queen's University Belfast. Iván’s research interests include AI approaches for decision making under uncertainty, consensus building, cooperative autonomous intelligent systems, collaborative filtering recommender systems, fuzzy preference aggregation and data fusion. Applications of his research include group recommender systems, disaster management, cybersecurity and energy planning. He has co-authored 12 publications in international journals and over 30 contributions to conferences, along with his recently published co-edited Springer book “Data Analytics and Decision Support for Cybersecurity”.

Date:  Wednesday, 25th October 2017

Time: 13:30 – 14:30

Location: B5 AMBS East

Title: Unsupervised Feature Selection in the Context of Online Bidding: A First Investigation

Presenter: Dr. Emiao Lu,  KTP Associate (Data scientist) at Dream Agility

E-mail: emiao.lu@manchester.ac.uk 

Abstract:

Online bidding system has gleaned increasing research interest with the rapid development of online advertising services. Such bidding systems aim to maximize advertisers’ profitability by optimally pricing online ads.  As an emerging research area, there has been limited work reported to tackle the challenge of optimizing online bidding. Here, we propose an innovative methodological framework that first partitions products into homogeneous groupings, where products fall into each cluster are used to construct a campaign. Subsequently, a suitable optimization algorithm is employed to each campaign to optimize the overall profitability of the cluster of products.As the initial stage of the research project, the first important issue related to the clustering of massive products is of selecting a meaningful subset of the original features, namely unsupervised feature selection. It is an essential step for improving the final performance and efficiency of clustering procedures in the presence of high-dimensional data. Here, we primarily focus on investigating the unsupervised feature selection problem in the context of online bidding where massive products are commonly present and required to be clustered. We attempt to compare the performance of some well-established feature selection methods, that are typically application independent, and application-based approaches that select the most promising features based on the profitability performance. By considering the problem-specific context during the feature selection step, we expect to discover features that are more valuable to the overall purpose of the application and thus lead to improved profitability.

Date:  Wednesday, 28th June 2017

Time: 13:00 – 14:00 with light lunch provided from 12:30   

Location: B4 AMBS East

Title: Analysing stochastic multi-criteria data by means of the empirical attainment function 

Presenter: Dr. Manuel López-IbáñezLecturer in the Decision and Cognitive Sciences Research Centre at the Alliance Manchester Business School, University of Manchester, UK http://lopez-ibanez.eu/

E-mail: manuel.lopez-ibanez@manchester.ac.uk

Abstract:

The attainment function is a generalisation of the cumulative distribution function for describing the statistical distribution of sets of mutually nondominated vectors. As such, it can be used to summarise and compare stochastic sets of multi-criteria alternatives or the output of multi-objective optimizers. There has been substantial theoretical work on the attainment function and its properties. In practice, the attainment function is estimated from samples of experimental data. This empirical attainment function (EAF) may used for statistical testing and graphical exploration thanks to existing software (http://lopez-ibanez.eu/eaftools). However, practical uses are often limited to bi-objective applications on one problem instance. There are both theoretical and practical challenges that need to be addressed to fully explore all the potential benefits of the EAF.

Date:  Wednesday, 14th June 2017

Time: 13:30 – 14:30 with light lunch provided from 13:00

Location: B7 AMBS East

Title: Condition Monitoring Makes the Energy Systems Smarter

Presenter: Dr. Xiandong MaEngineering Department, Lancaster University, UK http://www.lancaster.ac.uk/engineering/about-us/people/xiandong-ma

E-mail: xiandong.ma@lancaster.ac.uk

Abstract:

Monitoring and diagnostics of energy systems and their subsystems will play an increasing role in their competitive operation. Consolidated knowledge about their past and current conditions can be used to improve performance, efficiency, reliability and availability, thus enabling the optimal scheduling of maintenance activities and minimising the risk of costly unexpected failure during their service life. However, measurement signals are often non-stationary, stochastic and even more complex in a harsh environment like offshore wind farms. Furthermore, large volumes of data need to be processed and transmitted for further analysis, especially for continuous online monitoring. Minimising the number of sensors whilst still maintaining sufficient information to assess the system’s conditions is a critical concern for condition monitoring. Our research has been cantered on the study of data-driven models for effective condition monitoring using an intelligent and integrated approach. Through analysis of various signals, developing faults are diagnosed and prognosed well ahead of damage affecting the system. This talk will present the smart condition monitoring techniques and the associated instruments recently developed at Lancaster University.

Date:  Wednesday, 31st May 2017

Time: 13:00 – 14:00 with light lunch provided from 12:30

Location: B7 AMBS East

Title: How Marketing Vocabulary was evolving from 2005 to 2014? A content analysis based on statistical methods

Presenters: Dr. Igor BarahonaCONACYT Research FellowInstitute of Mathematics of the National Autonomous University of México (UNAM), Mexico https://www.researchgate.net/profile/Igor_Barahona

E-mail: igor@im.unam.mx

Abstract:

Here a collection of 1169 abstracts, which corresponds to articles that the Journal of Marketing Research has published from 2005 to 2014, are analysed under a novel approach. We apply several statistical methods, such as Principal Components Analysis and Correspondence Analysis to identify the way Marketing vocabulary is evolving. Similarly those articles that introduce new vocabulary are identified and the preferred words by authors are also detected.

In order to provide an easy-to-understand explanation, we provide our results graphically. A word-cloud with the most frequent words is given first. Secondly abstracts-words are represented on the factorial plane. Finally one representation of word-years allows us to detect changes on the vocabulary through the passing of time.

Date:  Wednesday, 17th May 2017

Time: 14:30 – 15:30 with light lunch provided from 14:00  

Location: B6 AMBS East

Title: Situational Awareness and Decision Making under Uncertainty for Autonomous Systems

Presenters: Prof. Wen-Hua ChenDepartment of Aeronautical and Automotive Engineering, Loughborough University, UK http://www.lboro.ac.uk/departments/aae/about/staff/professor-wen-hua-chen.html

E-mail: w.chen@lboro.ac.uk

Abstract:

Situational awareness and decision making are two core enabling functions for the safe operation of autonomous (aerial) vehicles and other autonomous systems. Due to the change of operation environment, limited sensory capability and accuracy, and possible change of intention of other objects, uncertainty widely exists in shared, public environments, and it is essential to cope with these kinds of uncertainty in information and environment in the development of these systems. Bayesian framework provides a generic and promising methodology in incorporating all the contextual knowledge, work model and historical data with real time senor measurements so to reduce the level of uncertainty in situational awareness and decision making. Particle filtering and Bayesian brief networks are adopted to facilitate the implementation of Bayesian framework. This talk overviews some relevant research activities currently taking place at Loughborough particularly on unmanned aircraft systems. Several autonomous functions for normal operation and contingency management (e.g. forced landing systems) of unmanned aircraft will be introduced.

Date: Wednesday, 26th April 2017

Time: 13:00 – 14:00 with light lunch provided from 12:30

Location: B6 AMBS East

Title: Strategies on improving maritime transportation safety of the Yangtze River

Presenters: Professor Xinping YanProfessor of Marine Engineering, Reliability and Safety in Wuhan University of TechnologyDirector of National Engineering Research Center for Water Transport Safety (WTS Center), China. http://wts.whut.edu.cn

E-mail: xpyan@whut.edu.cn

Abstract: 

With the fast development of the shipping industry, safety has been one of the top concerns for ships navigating in the Yangtze River. A lot of efforts have been done by researchers and maritime safety managers from technological or management perspectives. In this paper, the strategies on improving the maritime transportation safety of the Yangtze River are discussed and summarized. The navigation safety can be improved by using advanced technologies, applying the latest risk assessment theories and methodologies, or utilizing better safety management. Therefore, the strategies are composed of the four aspects, which are the more comprehensive data collection on ships and navigational environment, risk assessment and control based on statistics of the safety related data and on expert knowledge, modelling and simulations on emergency decision support, and maritime safety management systems. The work of the paper is meaningful for researchers and managers to get a full picture of the latest development in safety management and to make further efforts on enhancing navigation safety of the Yangtze River.

Date: Wednesday, 22nd March 2017

Time: 13:30 – 14:30 with light lunch provided from 13:00

Location: B6 AMBS East

Title: Roles of Multicriteria Decision Analysis in Public Sector Strategic Planning

Presenters: Professor Theo StewartEmeritus Professor of Statistical Sciences and Senior Research Scholar, University of Cape TownPart-time Professor of Decision Science, Alliance Manchester Business School, University of ManchesterEditor-in-Chief, Journal of Multi-Criteria Decision Analysis http://www.stats.uct.ac.za/stats/people/academic/stewart

E-mail: theodor.stewart@uct.ac.za

Abstract: 

The concepts developed in this presentation arose specifically in the context of national energy planning in developing countries, taking into consideration both reaction to and mitigation of climate change. The concepts undoubtedly apply equally to other strategic natural resource planning problems. We identify three phases in such strategic planning processes: an initial identification of courses of action that can be implemented; an assembly of such actions into portfolios that constitute potential policies; and the evaluation of such policies to provide final recommendations.Each phase can be viewed as a multiple criteria decision making problem in its own right, but different MCDA mechanisms will be appropriate to each. The first has a strong problem structuring element and discrete choice MCDA applied to a sorting problematique. The second is a multiobjective portfolio optimization problem, with the aim of generating a short-list for final consideration, within which we apply multiple reference point approaches. The third phase is again a discrete choice problem aimed at choice or ranking of alternatives, often in the presence of important qualitative criteria. We shall trace the development and integration of MCDA thinking through these three phases, and the need for backtracking at times to earlier phases. The approach will be illustrated by reference to earlier work in water resources planning, with some hypothetical extensions to create a clear numerical example.

Date:  Wednesday, 15th March 2017

Time: 13:30 – 14:30 with light lunch provided from 13:00    

Location: B2 AMBS East

Title: Robust Multiobjective Optimization for Decision Making under Uncertainty and Conflict

Presenters: Prof  Margaret M. WiecekDepartment of Mathematical Sciences, Clemson University, Clemson, SC, USA https://mthsc.clemson.edu/directory/view_person.py?person_id=89

E-mail: wmalgor@clemson.edu

Abstract:

Many real-life problems in engineering, business, and management are characterized by multiple, conflicting objectives, as well as the presence of uncertainty. The conflicting criteria originate from various ways to assess system performance and the multiplicity of decision makers, while uncertainty results from inaccurate or unknown data due to imperfect models and measurements, lack of knowledge, and volatility of the global environment.

In this talk, the deterministic approaches to uncertainty that are integrated with multiobjective optimization to address decision making under uncertainty and conflict are discussed. The approaches are based on robust optimization and parametric optimization, both developed for single-objective settings. Six sources of uncertainty are presented, and each type of uncertainty is placed in the multiobjective optimization problem (MOP), yielding several types of uncertain MOPs (UMOPs). Some of the sources are adopted from earlier studies in (single-objective) engineering optimization, while the others result from the multiobjective optimization modus operandi. The UMOP models are classified first according to the location of the uncertainty in their formulation, second with respect to the undertaken optimization approach, and third on the basis of the proposed definition of robust efficient solutions. The models are presented along with the accompanying results on solution concepts, properties, methods, and applications that are specific to each case.

Date:  Wednesday, 18th January 2017

Time: 13:30 – 14:30 (light lunch will be provided from 13:00)

Location: B7 AMBS East

Title: A year in the life of an applied statistician and possible MSc projects arising therefrom

Presenters: Professor Rose BakerEmeritus Professor of Applied Statistics, School of Business, University of Salfordhttp://www.salford.ac.uk/business-school/business-academics/rose-baker

E-mail: r.d.baker@salford.ac.uk

Abstract: 

The talk covers Rose’s recent research, focussing on work that could be developed into Msc projects for business studies. The research includes medical studies, with emphasis on meta-analysis, ranking players and teams in sports, study of lotteries like Euromillions, decision theory, the creation of new statistical distributions, and developing statistical tests. There are three main possibilities for projects: improving ranking methods in sport and marketing, developing a model of how juries reach decisions (deliberation), and developing a general 2-sample statistical test that could be useful in credit scoring. There are also other possibilities, such as further development of distributions for count data, and developing models for meta-analysis. Here, sports analytics is probably of most interest. Ranking players and teams is of great interest to the general public, while ranking players in football is useful for both managers and for opposition teams. Fortunately, sports analytics not only can be included under the ‘business’ heading, but also the techniques of ranking are transferable, for example to marketing.

Date: Wednesday, 7th December 2016

Time: 12:00 – 13:00 (light sandwich lunch will be provided from 11:30)

Location: B7, AMBS East

Title: Statistics in outcome prediction for trauma

Presenter: Dr Omar BouamraMedical Statistician at the Faculty of Biology, Medicine and Health, The University of Manchester http://www.manchester.ac.uk/research/omar.bouamra/personaldetails

E-mail: Omar.Bouamra@manchester.ac.uk

Abstract: 

Trauma is the leading cause of death worldwide amongst the young population (18 – 40 yrs).  The Truam Audit & Research Network (TARN) is the largest trauma registry in Europe (500k records), its role is to help emergency health systems to improve the care of their patients by providing them with useful information regarding their performance using different indicators and process measures.  I will be presenting the statistical methods used to derive hospitals outcome performance by means of a directly standardised statistic: Ws, that represents the excess survivor rate per 100. Outcome performance will be displayed using caterpillar and funnel plots where hospital outliers will be identified.  A follow-up graphical representation based on CUSUM technique and variable life adjusted display (VLAD) that helps identifying specific cases for review will be presented.

Date: Wednesday, 19th October 2016

Time: 12:00 – 13:00 (light sandwich lunch will be provided from 11:30)

Location: E21, AMBS East

Title: Solving multiobjective optimization problems by iteratively interacting with a decision maker

Presenter: Dr Jussi HakanenSenior researcher, PhDDept. of Mathematical Information Technology, University of Jyväskylä, Finland http://users.jyu.fi/~jhaka/en/

E-mail: jussi.hakanen@jyu.fi

Abstract: 

Typically, real-world optimization problems contain multiple conflicting objectives that need to be optimized simultaneously. These multiobjective optimization (MO) problems have several Pareto optimal (PO) solutions where none of the objectives can be improved without impairing some other objective. Mathematically all Pareto optimal solutions are equally good for the MO problem and further ordering them requires some additional information. Solving MO problems can mean different thing for different people: 1) identifying all PO solutions, 2) finding a representation for the Pareto optimal front, i.e., PO solutions in the objective space or 3) finding a most preferred solution for a human decision maker (DM) who can express preference information related to Pareto optimal solutions. During the last 50 years, many different types of methods have been developed for solving MO problems both in the fields of multiple criteria decision making (MCDM) and evolutionary multiobjective optimization (EMO). Typically, MCDM methods combine DM preferences in order to find a most preferred PO solution while the goal for population based EMO methods have been to find a representation of the whole Pareto front. MCDM methods typically work with one solution at a time, assume certain properties from the considered functions and are guaranteed to produce PO solutions. On the other hand, EMO methods are population based, don’t assume anything of the functions and don’t have any guarantee on Pareto optimality of the solutions produced. This presentation focuses on interactive MO methods where the most preferred solution is found by iteratively utilizing preferences of a decision maker. The interaction consists of phases where 1) decision maker evaluates computed (PO) solutions and expresses preferences on how to improve them, and 2) new (PO) solutions are computed using the preference information given. The background of interactive methods is coming from MCDM but, recently, the ideas have been utilized also in EMO. Examples of interactive methods and some applications are discussed.

Date: Thursday, 22nd September 2016

Time: 13:30 – 14:30 (light sandwich lunch will be provided from 13:00)              

Location: E21, AMBS East

Title: An Overview of Visualization Methods for Multiobjective Optimization

Presenter: Dr Tea TušarPostdoctoral Researcher at the Department of Intelligent Systems, Jožef Stefan Institute, Ljubljana, Slovenia

http://dis.ijs.si/tea/

E-mail: tea.tusar@ijs.si

Abstract:

When solving a multiobjective optimization problem, an algorithm commonly produces a set of mutually nondominated solutions that approximates the Pareto front for that problem. Visualization of such approximation sets is important for various purposes, for example, analysis of entire sets or inpidual solutions, decision support in interactive optimization, and analysis of algorithm performance. This seminar first introduces a methodology for evaluating and comparing visualization methods that uses two benchmark approximation sets and a list of desired properties for visualization methods (for example, preservation of the dominance relations among solutions, preservation of the distribution of solutions, robustness and scalability of the method, etc.) to analyze a visualization method based on its performance on the benchmark sets. It then uses this methodology to assess 20 visualization methods used in multiobjective optimization and present their advantages and limitations. 
Date: Thursday, 28th July 2016

Time: 13:00 – 14:00 (light sandwich lunch will be provided from 12:30)              

Location: B7, AMBS East

Title: Intelligent Systems for Smart Cities

Presenter: Professor Enrique Alba

Professor of Computer Science, University of Málaga, Spain

http://www.lcc.uma.es/~eat/

E-mail: eat@lcc.uma.es

Abstract:

The concept of Smart Cities can be understood as a holistic approach to improve the level of development and management of the city in a broad range of services by using information and communication technologies.

It is common to recognize six axes of work in them: i) Smart Economy, ii) Smart People, iii) Smart Governance, iv) Smart Mobility, v) Smart Environment, and vi) Smart Living. In this tutorial we first focus on a capital issue: smart mobility. European citizens and economic actors need a transport system which provides them with seamless, high-quality door-to-door mobility. At the same time, the adverse effects of transport on the climate, the environment and human health need to be reduced. We will show many new systems based in the use of bio-inspired techniques to ease the road traffic flow in the city, as well as allowing a customized smooth experience for travellers (private and public transport).

This talk will then discuss on potential applications of intelligent systems for energy (like adaptive lighting in streets), environmental applications (like mobile sensors for air pollution), smart building (intelligent design), and several other applications linked to smart living, tourism, and smart municipal governance.

Date: Wednesday, 27th July 2016

Time: 13:00 – 14:00, light sandwich lunch will be provided from 12:30

Location: B7 AMBS East

Title: The role and impact of comparison websites on the consumer search process in the US and German airline markets

Presenters: Professor Christopher P. Holland

Senior Lecturer, Department of Automatic Control and Systems Engineering, University of Sheffield

https://www.sheffield.ac.uk/acse/staff/rp

E-mail: r.purshouse@sheffield.ac.uk

Abstract: The paper examines how consumers search for airline tickets based on a comparative analysis of the US and German markets. Data from comScore is analysed using an innovative application of set theory. ComScore is a leading commercial provider of business intelligence and consumer analytics based on its worldwide panel of two million online users. The search process is modelled using the concept of the consideration set based on primary search with the airline websites and secondly by the use of online travel agents and meta-search engines, which are termed comparison websites. Three generic search models are proposed: (1) primary search with airline websites only; (2) search of comparison websites only; (3) a combination of primary search and comparison websites. Each generic search model accounts for a significant proportion of overall users in both markets. The consideration sets are 2.58 in Germany and 2.74 in the United States. It is shown that the use of comparison websites significantly increases the propensity to conduct additional primary search based on analysis of all major airline pairs in both markets. The theoretical and managerial implications of the results are described and future research opportunities are outlined.

Date: Wednesday, 29th June 2016

Time: 13:00 – 14:00 (light sandwich lunch will be provided from 12:30)

Location: B8, AMBS East

Title: Towards asynchronous distributed optimization and decision support for complex engineered productsPresenters: Dr Robin PurshouseSenior Lecturer, Department of Automatic Control and Systems Engineering, University of Sheffield https://www.sheffield.ac.uk/acse/staff/rp

E-mail: r.purshouse@sheffield.ac.uk

Abstract:The design of complex engineered products, such as modern automotive vehicles, in principle offers good opportunities for optimization and decision support technologies to deliver benefit. However, the complex socio-technical environment in which these products are created presents barriers to the successful deployment of existing methods. This seminar will review existing attempts to account for the complexities of the engineering design environment, including frameworks from the field of multi-disciplinary design optimization (MDO). It will also highlight the key challenges as yet unaddressed by the MDO community; specifically: (1) how to handle the asynchronous distributed nature of the engineering design environment to ensure right-first-time design; (2) how to allocate resources to compromise-seeking activities in an environment of shared design variables and conflicting product requirements. Based on on-going research, the seminar will sketch a new, distributed optimization architecture and provide two realisations based on alternative decision-making approaches within the product creation process.

Date: Wednesday, 1st June 2016

Time: 14:00 – 15:00 (light sandwich lunch will be provided from 13:30)

Location: B7, AMBS East

Title: Information privacy leakage concerns: A case study of Facebook social advertising

Presenter: Dr Maged AliSenior Lecturer in Digital Marketing, Essex Business School, University of Essex, Essex, UK https://www.essex.ac.uk/ebs/staff/profile.aspx?ID=4947

E-mail: mailto:maaali@essex.ac.uk

Abstract:The global popularity of Facebook offers a hugely tempting resource for organisations engaged in online business. The personal data willingly shared between online friends’ networks intuitively appears to be a natural extension of current advertising strategies such as word-of-mouth and viral marketing. However, commercial use of such personal information has provoked outrage amongst Facebook users. To date, little is known about how SNS users perceive such leakage of privacy. So a greater understanding of the form and nature of SNS users’ concerns about privacy leakage would help to formulate best practice guidelines for organisations regarding the use of Facebook users’ data in social media marketing. Given the fluid, context-dependent and temporal nature of privacy, a longitudinal case study representing the launch of Facebook’s social Ads programme was conducted to investigate the phenomenon of privacy leakage within its real-life setting. A qualitative user blogs commentary was collected over three years during the two-stage launch of the social Ads programme. Grounded theory was used to analyse users’ blog postings. The resulting model of privacy leakage concerns contributes to improve our understanding of the nature, scope and form of the core privacy leakage concerns of Facebook users. The current research has economic and political implications.

Date: Wednesday, 18th May 2016

Time: 14:00 – 15:00 (light sandwich lunch will be provided from 13:30)

Location: Penthouse, AMBS West

Title: Forecasting for Big data: does sub-optimality matter?

Presenter: Professor Konstantinos I. NikolopoulosChair in Decision Sciences, Bangor Business School, Prifysgol Bangor University, Bangor, Wales https://www.bangor.ac.uk/business/staff/kostas_nikolopoulos.php.en

E-mail: kostas@bangor.ac.uk

Abstract:With all the plethora of sources, high frequencies and vast volumes of data - quantitative as in panel data as well as qualitative eg semantically filtered and sequentially quantified tweets on product's quality, the life of the forecaster has not become easier at all. You may have more information available (actually it is just data...) but nobody knows exactly how to employ all that information for improving the accuracy and efficiency of the forecasting function in 21-st century organisations. On the contrary very often forecasters start analysing irrelevant data by using more data than what they should; and they end up getting computationally intensive forecasting algorithms and the computers servers that run them to their knees... on the quest of optimal parameter setting of forecasting models so as to get more accurate forecasts. But is this the case? Is this is obsession to optimality always bare the respective fruits or we actually do lose too much time on the quest of it? And could we better off maybe by aiming for much faster and robust systems that would aim for suboptimal forecasting solutions not deteriorating accuracy that much, at least not that much to jeopardise the efficiency of the systems under use. This study throws light to that end through an extensive empirical investigation and a series of deductive arguments, all in favour of chasing suboptimal forecasting model parameters (under constraints of course...)

Date:  Wednesday, 27th April 2016Time: 13:00 – 14:00 (light sandwich lunch will be provided from 12:30)

Location: B7, AMBS East

Title: Integrating Preference Information into Evolutionary Multi-objective Optimization

Presenter: Professor Juergen BrankeProfessor of Operational Research & Systems, Operational Research & Management Sciences Group, Warwick Business School, University of Warwick, U Khttp://www.wbs.ac.uk/about/person/juergen-branke/

E-mail: Juergen.Branke@wbs.ac.uk 

Abstract:Many practical optimization problems require the consideration of multiple, conflicting objectives. In such cases, usually no single optimal solution exists. Instead, there is a set of so-called efficient or Pareto-optimal solutions with different trade-offs. Evolutionary algorithms, i.e., heuristics inspired by natural evolution, have gained increasing popularity for such multi-objective problems. Since they work on a population of solutions, they can be used to simultaneously search for a well-distributed set of Pareto-optimal solutions in a single run. This provides the decision maker with a set of alternatives to choose from.

This talk will start with a brief introduction to evolutionary multiobjective optimisation, and then discuss why and how the decision maker’s preferences should be incorporated, either a priori or during the optimisation. A number of different ways to incorporate user preferences are presented, including some recent approaches that learn the user preferences from pairwise comparisons during the run.

Date: Wednesday, 20th April 2016

Time: 15:00 – 16:00

Location: Alumni Common Room, AMBS West

Title: Markowitz Portfolio Selection – with a Third Criterion?

Presenter: Professor Ralph E. SteuerCharles S. Sanford, Sr. Chair of Business, Department of Finance, Terry College of Business, University of Georgia, USAhttp://people.terry.uga.edu/rsteuer/profile.htm

E-mail: rsteuer@uga.edu

Abstract:In 1952 Markowitz introduced the mean-variance efficient frontier to finance. While mean-variance is still the predominant model in portfolio selection, it has endured many criticisms. One of its most persistent has been that it does not allow for additional criteria. The difficulty is that with additional criteria, the efficient frontier becomes a surface. With results on how to compute surfaces, a look is taken at the exact translation of Markowitz’s risk-return (bi-criterion) approach to tri-criteria when the extra objective is linear (such as for social responsibility and dividends). The case of when the third criterion is quadratic is also mentioned. With the geometry of the translation playing a major role, many graphs are used to illustrate.

Date: Wednesday, 20th January 2016

Time: 13:30 – 14:30

Location: B7 AMBS East

Title: Data envelopment analysis (DEA) and DEA into Policy Analysis - Recent research progress at Manchester Business School

Presenter: Dr. Guoliang YangAssociate Professor, Institute of Policy and Management, Chinese Academy of SciencesVisiting Scholar at AMBSE-mail: glyang@casipm.ac.cn

Abstract:This presentation aims to introduce the recent research progress of Dr. Guoliang Yang at Manchester Business School as an academic visitor. It is two-fold:- The first part of the presentation is to report a research on two-stage DEA. Traditional DEA treats the production as a "black box". Recently more and more researchers in this field intend to go inside the "black box" and the internal structure of DMUs. This paper aims to propose a new approach to decompose the overall efficiency of two-stage DEA model into different stage separately. In this approach we use a minimax model as the equalizer to set the equal priority on two stages. This new approach satisfy that the overall efficiency is equal to the combination of its weighted decompositions.- In the second part, we analyze the impacts of the Higher Education Reforms implemented in the 2000s on the optimal size of universities in Germany. We hypothesize that smaller universities were better able to adapt to the Higher Education Reforms, triggering a decline in the optimal size of universities in the reform period. Using a 12-year panel data set on the inputs and outputs of German universities, we find a tremendous decrease in optimal university size in terms of students. We find that this drop can be mainly attributed to the year 2002, which coincides with the implementation of the Bologna Reforms, suggesting that this was the main trigger. Furthermore, this decline correlates with a relative increase of the overheads in larger universities, providing evidence in favor of the hypothesis that the decline was driven by relatively greater increases of administration overheads in larger universities.

Date: Wednesday, 9th December 2015

Time: 13:00 – 14:00

Location: B1 AMBS East

Title: Multicriteria decision analysis for assessing environmental risks: reflections on the UK's Committee on Radioactive Waste Management

Presenter: Professor Alec MortonProfessor of Management Science, University of Strathclyde, GlasgowVisiting Professor, University of Science and Technology of China, Hefeihttp://www.strath.ac.uk/staff/mortonalecprof/E-mail: alec.morton@strath.ac.uk

Abstract:This talk will have a practical and a methodological component. The practical part will concern a experience of the successful use of a multi criteria decision analysis approach to support policy making about radioactive waste in the United Kingdom. In the theory part I will discuss the strengths and limitations of using an additive multicriteria model. I will present some analytic machinery which can be used to handle decision maker preferences when the underlying value function may not be additive.

Date:  Wednesday, 2nd December 2015

Time: 15:00 – 16:00

Location: B7 AMBS East

Title: Automatic algorithm configuration with irace: new developments, challenges and applications

Presenter: Leslie Pérez CáceresPhD student at IRIDIA, Université Libre de Bruxelles http://iridia.ulb.ac.be/~lperez/info.htmlE-mail: leslie.perez.caceres@ulb.ac.be

Abstract:The parameters values  of an algorithm define its behavior when solving a problem, finding good sets of parameter values is a crucial task given its great impact on performance. Algorithm configuration, also called parameter tuning, is the process of finding sets of good parameters values and is often performed manually or not performed at all. Automatizing the configuration process allows developers and users to save time and resources normally used to configure algorithms. The automatic configuration of algorithms has been applied successfully in several areas, academic and industrial, to configure optimization algorithms, hyper-parameters of machine learning models, simulation parameters,  robotic systems and compiler optimization options.

This talk will introduce the automatic configuration of algorithms using irace, an open source automatic configuration algorithm developed at IRIDIA based on iterated F-race.  We review the applications of irace and we introduce a new version of irace, elitist irace, the ongoing developments  and the challenges regarding automatic configuration that will be approached in future research.

Date:  Thursday, 23rd April 2015

Time: 12:00 – 13:30                                                      

Location: 5.1Crawford House (5th floor,front to Alliance MBS East)

Title: Decision making and fuzzy logic in brazil: Desire, Pricing, Evaluation and Perception

Presenters:   Ilan Chamovitz,D.Sc;   Fabio Krykhtine,M.Sc;   Luís Raymundo,M.Sc;  Antonio Carlos Morim,M.Sc;  Guilherme Martins,M.Sc.;  Paulo Reis Filho, D.Sc.

LabFuzzy - Production Engineering Program, COPPE, Rio de Janeiro Federal University (UFRJ),

Centro de Tecnologia, bloco F sala F110, Cidade Universitária - Ilha do Fundão, CEP 21.949-900 - Rio de Janeiro - RJ

http://www.api.adm.br/fuzzy

E-mail: ilan@api.adm.br  

 Date:  Wednesday, 18th March 2015

Time: 12:45 – 14:00                                                      

Location: 3.103 Alliance MBS West

Title: Sensitivity Methods for the Management Sciences

Presenter: Professor Emanuele Borgonovo

Department of Decision Sciences, Bocconi University, 20136, Via Roentgen 1, Milano, Italy.

http://faculty.unibocconi.eu/emanueleborgonovo

E-mail: emanuele.borgonovo@uni-bocconi.it

Date:  Wednesday, 4th February 2015

Time: 13:30 – 14:30

Location: B7, Alliance MBS East

Kostas Nikolopoulos presents @ INFORMS annual conference - San Fransisco on 12 Nov 2014 at 16:30 : “DIY Forecasting: Judgment, Models and Judgmental Model Selection" 

Title: 

DYI Forecasting: Judgment, Models and Judgmental Model Selection

 

Presenting Author: 

Konstantinos Nikolopoulos,Bangor Business School, Bangor University, College Road, Bangor LL57 2DG, United Kingdom, k.nikolopoulos@bangor.ac.uk

 

Co-Author: 

Nikolaos Kourentzes,Lancaster University, The Management School, Lancaster University, Lancaster LA1 4YX, United Kingdom, n.kourentzes@lancaster.ac.uk

 

 

Fotios Petropoulos,Cardiff Business School, Cardiff University, Aberconway Building, Cardiff CF10 3EU, United Kingdom, PetropoulosF@cardiff.ac.uk
 

Abstract: 

In this paper we explore how judgment can be used to improve statistical model selection for forecasting. We investigate the performance of various judgmental model selection methodologies against the benchmark statistical one, based on information criteria. We evaluate the performance of experts in terms of selecting the best model and forecasting performance, identifying major improvements. We examine how to extend statistical model selection to incorporate additional insights from experts.
Date:  Wednesday, 3rd December 2014

Time: 13:30 – 14:30                                                      

Location: B1 Alliance MBS East

Title: Corroborative and DisCorroborative Evidential Record Aggregation

Authors/Presenters: Dr. Rebecca Copeland

Kenilworth, Warwickshire CV8 2QF, UK

Mobile +44 7879 490494 Desk +44 1926 777280

Email: rebecca.copeland@coreviewpoint.com

Date:  Wednesday, 5th November 2014 

Time: 13:30 – 14:30     

Location: B1 Alliance MBS East

Title: Generalising Bayesian Inference to Evidential Reasoning with Sample Data

Authors/Presenters: Prof. Jian-Bo Yang and Dong-Ling Xu

http://personal.mbs.ac.uk/jyang                                        

Abstract: In this presentation, the relationship between Bayes’ rule and the Evidential Reasoning (ER) rule is explored. The ER rule has been uncovered recently for inference with multiple pieces of uncertain evidence profiled as a belief distribution and takes Dempster’s rule in the evidence theory as a special case. After a brief introduction to the ER rule, the conditions under which Bayes’ rule becomes a special case of the ER rule are established. The main findings include that the normalisation of likelihoods in Bayesian paradigm results in the degrees of belief in the ER paradigm. This leads to ER-based symmetrical probabilistic inference, where evidence can be generated from sample data and profiled in the same format of belief (probability) distribution, thereby stimulating an avenue of research and application in data-driven evidential reasoning. Numerical examples are examined to demonstrate the findings and their potential applications in probabilistic inference. In this presentation, it will also be shown that these findings enable the generalisation of Bayesian inference to evidential reasoning with ambiguous probability information of different weights and reliabilities.

Date:  Friday, 15th  August 2014

Time: 14:00 – 15:00                                                      

Location: B8 Alliance MBS East

Title: Introduction to Transport Science and Technology in Wuhan University of Technology

Presenter:  Xinping Yan, PhD

Professor and Director of Engineering Centre for Safety (Ministry of Education)

Vice President of Wuhan University of Technology

1040 Heping Avenue, Wuhan, 430063, P. R. China.  Email: xpyan@whut.edu.cn

Bio:  Prof. Yan received his BSc in Marine Mechanical Engineering from Wuhan University of Water Transportation Engineering, China in 1982 and his Master from the same University in 1987, received his PhD in Mechanical Engineering from Xi’an Jiaotong University, China in 1997. He has been devoted to ship reliability research for the past 20 years. Currently as the principal investigator he holds 4 externally funded grants (2 by the NSFC and 1 by China Maritime Safety Administration). He has completed supervision of 27 PhD and 5 postdoctoral researchers in the area of transportation safety. Currently, Prof. Yan’s publications include 34 SCI cited journal papers, over 100 other journal papers, and over 100 conference papers. Prof. Yan’s major research interests are in the areas of condition monitoring and fault diagnosis, marine system design and control, tribology and its industrial application, intelligent transport system, and maritime education etc. He awarded as the Distinguished Visiting Fellowship Award from the Royal Academy of Engineering in 2014.

Prof. Yan has been collaborating with many industrial companies and regulatory bodies including China Ocean Shipping COmpany (COSCO) and China Maritime Safety Administration (MSA). He is the editor-in-chief of a Chinese journal, Journal of Transport Information and Safety, and an editorial board member/an associate editor of 5 international/national journals including Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment.       

 

Weekly workshops for DCS PhD researchers 09/2013 – 07/2014 

Date:  Wednesday, 28th May 2014

Time: 13:30 – 14:30                                                      

Location: B5 Alliance MBS East

Title: Artificial Intelligence in Operations

Presenter:  Professor Khairy Kobbacy

Abstract:  The use of Artificial Intelligence (AI) in Operations Management, with its ability to evolve solutions, handle uncertainty and perform optimisation continues to be a major field of research.  The growing body of publications over the last two decades means that it can be difficult to keep track of what has been done previously, what has worked, and what really needs to be addressed.   This seminar reviews the AI applications in operations and presents the results of a recent survey of the use of AI in operations management covering the period 2009-2013 including  the key research themes, trends and directions of research. The seminar will also compare the results of this survey with those of previous surveys carried out since 1997 with the aim of identifying longer term trends and developments.

The  work  presented in this seminar is based on continuous collaboration between Khairy Kobbacy and Sunil Vadera of Salford University in the area of AI in Operations which resulted in the publication of several review papers in leading  international journals and  involved the organisation of five European conferences, all sponsored by the UK Operational Research Society. They also were Guest Editors of several special issues of Journals  in this area.

Date:  Wednesday, 7th May 2014

Time: 13:30 – 14:30                                                      

Location: B5 Alliance MBS East

Title: Approaches on decision making and priority setting in R&D and innovation investments

Presenter:  Professor Sergio Salles-Filho

University of Campinas – UNICAMP, Brazil, Manchester Institute of Innovation Research

Abstract: The presentation will be focused on the contributions of qualitative and quantitative approaches and methodologies (and on the possibilities of combining them) to deal with prioritization and decision making problems in R&D and Innovation activities. In R&D and innovation arena the basic information about a given idea, project or even about an investment in new products and processes is not yet available. There is no previous experience to be based on (except for imitation). From the point of view of a decision maker (an entrepreneur), this means that he or she will have to deal with partial knowledge and information, and that they will have to calculate using more or less sophisticated tools under a bounded rationality which implies their decisions will be made based on beliefs that are yet to be tested.

We will present some experiences of combining qualitative and quantitative approaches (expert opinion, multicriteria analysis, optimization methods) in priority setting and in decision making in R&D and Innovation projects in different sectors in Brazil. We look forward to exchange our findings and to explore new possibilities on this domain.

Date:  Wednesday, 26th February 2014

Time: 13:30 – 14:30                                                      

Location: 10.09 (Harold Hankins)

Title: Adaptive Computer Mediated Communication

Presenter:  Dr. Oscar De bruijn

http://www.manchester.ac.uk/research/o.de-bruijn/

Abstract: The notion of Common Ground has been a prominent theoretical construct in the study of computer-mediated communication (CMC) for well over twenty years. However, so far this has not lead to a coherent theory that both explains and accurately predicts people’s CMC behaviour. In this seminar I will outline a new project aimed at advancing the theory of common ground by studying how participants in CMC rationally adapt their language use in order to gain maximum utility. As we will see, this requires conceptualisation of CMC behaviour as a problem of decision making under uncertainty.  I will present the results of a preliminary study and outline our plans for future research.

Date:  Wednesday, 5th February 2014

Time: 13:30 – 14:30                                                      

Location: B7, Alliance MBS East

Title: Forecasting Black (& White) Swans...

Presenter: Konstantinos Nikolopoulos (a), Aris A. Syntetos (b)

Bangor Business School, Bangor University, Bangor, Gwynedd, LL57 2DG, Wales, U.K. ; k.nikolopoulos@bangor.ac.uk

http://www.bangor.ac.uk/business/staff/kostas_nikolopoulos.php.en

Abstract: “Forecasting White Swans” is not trivial, but at least there is a quite advanced arsenal –in the form of advanced mathematical forecasting models- that we may employ so as to accurately forecast phenomena that tend to be observable at regular frequencies. Call it time-series models, econometric models, computational intensive approaches as in Artificial Neural Networks … there is always a way to get a fair extrapolation of what is going to happen either in the form of a point forecast, or a density forecast: the latter being comprised of a prediction interval and a level of confidence associated with your belief that the forecasted value will actually lay in the forecasted range. However when it comes to „Black Swans‟ then we have a whole new level of a game - much harder: the question now becomes from “how many White Swans we will see tomorrow?” to “when the next Black Swan will be seen?” and “if so… will he be alone this time…?”. And in the lack of sufficient quantitative information judgmental forecasting approaches may have to be used this time. The story of this paper unfolds by adopting a technical approach on potentially useful mathematical OR/MS tools and techniques that could help us determine the forecasting horizon of our problem, that is the period ahead that we will reasonable expect at least one “Black Swan‟ to appear.

Keywords: Intermittent demand; forecasting; decomposition; baseline; extremes
Date:  Wednesday, 11th December 2013

Time: 13:30 – 14:30                                                      

Location: B7, Alliance MBS East

Title: From Bayesian Inference to Evidential Reasoning for Decision Making under Uncertainty

Presenter: Professor Jian-Bo Yang and Professor Dong-Ling Xu

Decision and Cognitive Sciences Research Centre

Alliance Manchester Business School, The University of Manchester, Manchester M15 6PB, UK

jian-bo.yang@manchester.ac.uk, l.xu@manchester.ac.uk

www.personal.mbs.ac.uk/jbyang, www.mbs.ac.uk/research/people/profiles/lxu

Abstract: This presentation is intended to introduce a unique Evidential Reasoning (ER) rule that can be used to combine conjunctively multiple pieces of independent evidence with various weights and reliabilities. The ER rule has a wide range of applications such as decision analysis under uncertainty, multiple criteria decision analysis, risk analysis, information fusion, medical diagnosis and fault diagnosis. The presentation is first focused on discussing prior knowledge including Bayesian inference and Dempster- Shafer theory of evidence, in particular Dempster’s rule of evidence combination. The necessity of taking into account the weight and reliability of evidence is discussed, as well as the concept of the degree of support for hypotheses from evidence. The novel concept of Weighted Belief Distribution with Reliability (WBDR) is then introduced to characterise evidence equivalently in complement of Belief Distribution (BD) introduced in Dempster-Shafer theory. The implementation of the orthogonal sum operation on WBDRs leads to the establishment of the new ER rule, which constitutes a generic conjunctive probabilistic reasoning process, or a generalised Bayesian inference process. The ER rule takes the original ER algorithm as a special case when the reliability of evidence is equal to its weight and the weights of all pieces of evidence are normalised. The ER rule also takes Dempster’s rule as a special case when each piece of evidence is fully reliable, with the latter completing and enhancing the former by identifying how to combine pieces of fully reliable evidence that are highly or completely conflicting through a so-called reliability perturbation analysis. The ER rule holds interesting properties such as being associative and commutative, so it can be used to combine multiple pieces of evidence in any order without changing the results of combination. Examples are discussed to help illustrate the main ideas throughout the presentation.

Date:  Friday, 22nd November 2013

Time: 13:30 – 14:30                                                      

Location: B3 Alliance MBS East

Title: Maritime Risk Assessment & Anti-collision Decision Making

Presenter: Di Zhang, PhD

Assistant Professor

Intelligent Transport Systems Research Center, Wuhan University of Technology, 1040 Heping Avenue, Wuhan, 430063, P. R.China, fred.zhangdi@gmail.com

Abstract: Waterway transportation plays an essential role for economic development. China Maritime Safety Administration (MSA) shoulders the responsibility to maintain waterway transportation safety and efficiency and to avoid environmental contamination. The first part of this talk focuses on the assessment of MSA performance in term of safety with Belief Rule-base (BRB) methodology. A generalization of traditional BRB theories, which is called G-BRB, is introduced. G-BRB focuses on the fact that experts’ subjective standards on one quantitative attribute may be different. The qualitative data of safety and cost attributes from expert questionnaires is transformed into antecedent belief structure (A-BS), which can well reflect the distinctions among experts’knowledge. The proposed method is used in a case study to assess the performance of one MSA in China during the year 2007–2011.

Among maritime accidents, collision is one of the most common types of accident sometimes causing enormous economic loss and serious casualties. In fact, a lot of collisions happen because the navigators of the ships misunderstand each other’s intention. Anti-collision support system is one of the most effective ways to reduce such failures. In the second part of the talk, a distributed anti-collision procedure under multi-ship situation is introduced. Each ship makes decisions by their own observations on other ships’ behaviors or actions to keep clear of all the ships that it should give way to. The procedure is carried out in a real-time mode. In order to promote the reliability, not only the situation that all the involved ships comply with the requirements of the International Regulations for Preventing Collision at Sea (COLREGs) is considered, but also the decision making for stand-on ship given that any give-way ship violates the rules is studied. 

Date:  Wednesday, 16th October 2013

Time: 13:30 – 14:30                                                      

Location: B1 Alliance MBS East

Title: Group Role Assignment and Transfer

Presenter: Haibin Zhu, PhD

Professor and Coordinator, Computer Science Program

Director, Collaborative Systems Laboratory, Nipissing University, 100 College Dr., North Bay, ON P1B 8L7, Canada, haibinz@nipissingu.ca

http://www.nipissingu.ca/faculty/haibinz

Abstract: Role-Based Collaboration (RBC) has been proposed as an emerging and promising approach to facilitating collaboration. It utilizes roles as underlying mechanisms to support collaboration by taking advantages of roles. It is divided into several phases: role negotiation, role assignment, and role play.

Role assignment can be categorized into three phases: agent evaluation, group role assignment, and role transfer. Agent evaluation rates the qualification of an agent for a role. It requires a check on the capabilities, experiences, and credits of agents based on role specifications. Qualifications are the basic requirements for possible role-related activities. It is a fundamental yet difficult problem that requires advanced methodologies, such as information classification, data mining, pattern search, and match. Group role assignment initiates a group by assigning roles to its members or agents to achieve its highest performance. Role transfer (also called dynamic role assignment) re-assigns roles to agents or transfers agent roles to meet the requirement of the system changes.

The availability of a group is highly dependent on its structure. To keep a group available, role transfer is required when the situation of a group changes. To keep high performance of a group, dynamic role assignment is required. Both problems are among the optimization problems, and  require careful plans and arrangements. The solutions of the above problems can be widely applied into many fields, such as, team work, Grid computing, sensor networking, engineering, organization, training, and management.

This talk introduces the concepts of Role-Based Collaboration, explains the benefits of role mechanisms, illustrates the RBC model E-CARGO, clarifies the group role transfer and assignment problems by examples and formalizations, discusses the solutions and presents newest application case studies.

Date:  Wednesday, 12th June 2013

Time: 13:00 – 14:30                                                      

Location: B5 Alliance MBS East

Title: Modelling Implicit Attitudes using Linguistic Information from the World Wide Web

Presenter: Dr Dermot Lynott, Alliance MBS

Abstract: Psychologists have long used implicit measures of judgement to uncover people's attitudes, particularly for sensitive topics where people are less likely to provide honest responses to explicit questions.  Over the last 15 years, the implicit association test (IAT) has proved one of the most popular methods of establishing whether people display a bias towards or against particular target concepts in many domains, including racism, gender bias, politics and religion. Importantly, the bias people display is also predictive of their decision making in related domains (e.g., racial prejudice predicting voting patterns). In previous work, we employed measures of linguistic distributional information (i.e., N-gram frequencies across a large corpus representing a snapshot of the world wide web) to model response patterns in language comprehension and conceptual processing tasks (e.g., conceptual combination, property verification).  In this work, we consider whether the patterns of bias uncovered by the IAT might also be predictable using the linguistic distributional characteristics of the target concepts (e.g., race) and associated attributes (e.g., positive and negative features).  

Title: Specialised Knowledge Visualization for Decision Making in Terminology Management

Presenter: Dr Juan Antonio Prieto Velasco, University Pablo de Olavide, Seville, Spain

Visiting academic researcher, Alliance Manchester Business School, University of Manchester

Abstract: Terminology management can be regarded as any deliberate manipulation of terminological information. Selecting appropriate terms, representing and describing pertinent concepts have often been a major concern for both terminologists and translators, who manipulate information in order to construct meaning in terminological databases and in the texts they have to translate. Specialised knowledge visualisation (SKV) is a recent approach which accounts for the embodied nature of specialised concepts and the multimodal representation of knowledge in terminological databases. Although pictorial representations can completely assume the function of both terms and definitions, one issue still remains unclear: what are the principles guiding the selection of images to depict specialized concepts? Decision making is therefore crucial to choose those images which best depict a given concept on the basis of the user’s prior knowledge, the informational content of images, subject-field conventions, etc. In this talk, we describe our research within SKV aimed at identifying a principled selection of images for their inclusion in terminological databases.

Date and Time: Wednesday 8th May 2013, 14:15 – 15:15

Location: B5 Alliance MBS East

Title: PowerMeeting: A Web-Browser-Based Group Decision Support System

Speaker: Dr Weigang Wang, Alliance MBS

Abstract: Web based group decision support, especially whose based on more sophisticated decision making methods, faces many challenges.  This work tries to promote wide participation in and better understanding on group decision making activities using various decision making methods, and  to facilitate a session chair led group decision making process, so as to help teams make well-informed, rational decisions in a well-coordinated group process.  In this talk, the challenges and the PowerMeeting approach to addressing them will be presented.  The PowerMeeting system and more details on it can be accessed at powermeeting.co.uk

Date:  Monday, 11th March 2013

Time: 13:30 – 15:00                                                      

Location: B3 Alliance MBS East

Title: TEI@I Methodology with Applications to Economic Forecasting

Presenter: Prof. Shouyang Wang

Abstract: In this talk, a new methodology -- TEI@I methology for conplex systems analysis is introduced. Some applications in economic forecasting are used to illustrate the methodology.

Shouyang WANG, TWAS Fellow Elected 2011, Membership: 10 –Social and Economic Sciences

Biographical data: Born in Jiangsu, China on 20 July 1958. Wang is professor at the Academy of Mathematics and Systems Science of Chinese Academy of Sciences (CAS) in Beijing, and the chief economist and the founding director of the Center for Forecasting Science of CAS. He obtained his PhD in Operations Research in 1986 at the Institute of Systems Science of CAS. He is the president of International Society of Knowledge and Systems Sciences, the president of China Society of Systems Engineering, a vice president of Chinese Academy of Management and an advisor of several ministries of China. He was awarded many prizes and awards including Research Fellowship of the Dutch Royal Academy of Sciences in 1987-1988, the Croucher Foundation Fellowship of Hong Kong in 1993, the CAS Prize for Young Scientists in 1993, the China Science and Technology Prize for Young Scientists in 1994, Japan Science Promotion Society Fellowship in 1994, The CAS 100 Talent Program Scientist in 1996, Foreign Professorship of Tsukuba University in Japan in 2000-2001 and 2003-2004, the NSFC Fund for Outstanding Scientists in 2002, Green Group Award for Business Intelligence and Computational Economics in Krakow, Poland in 2007, The President Award of International Society of Multiple Criteria Decision Making in 2007 and Fudan Premium Prize of Management in 2008. He was elected as an academician of International Academy of Systems and Cybernetics in Wien in 2010.

Research areas: Economic analysis and forecasting; game theory and conflict resolution; decision analysis and decision support systems; policy studies.

Address: Academy of Mathematics and Systems Science, Chinese Academy of Sciences, No. 55, Zhongguancun East Road, Haidian District, Beijing, 100190, China

Phone: (+86 10) 6265-1375, Fax: (+86 10) 6256-8364, E-mail: sywang@amss.ac.cnhttp://www.amss.ac.cn/~sywang

Cost IC0602 International Doctoral School

Applying Decision Analysis to Real Problems

Alliance Manchester Business School, 10th to 13th April 2011

http://research.mbs.ac.uk/decision-science/Newsandevents.aspx

Download online wordfile: http://research.mbs.ac.uk/decision-science/Portals/0/docs/IC0602_Doctoral_School_MBS.docx

Introduction

Alliance Manchester Business School will be holding a doctoral school in Manchester from Sunday 10th to Wednesday 13th April 2011 to explore the processes of applying decision analysis in practice.  Attendees will be introduced to some theory, particularly behavioural and similar theories which are often absent from mathematical and algorithmic presentations of decision analysis; but the majority of the course will focus on practical exercises and case studies.  Our aim will be to introduce participants to the range of skills that real application requires.

Content of the Doctoral School

The Doctoral School will offer the following lectures and activities

  • Introductory session surveying decision analysis and introducing the terminology and models to be used in the School;
  • Lectures including demonstrations and break-out sessions on:
  • Case studies of applications of decision analysis in domains such as
  • Supply chain risk;

The students will be provided with notes, copies of slides, etc.  A discussion web-site will be established during the event and run in the weeks after it to continue discussion and share experiences in applying the ideas back in their research institutes and universities.

Outline Timetable

Sunday April 10th 2011

14.00 – 15.00

Introduction.  Getting to know each other.

15.00 – 15.30

Tea

15.30 – 17.30

Lecture: What is decision analysis: a brief overview illustrated by case vignettes.

 

Wine and light buffet, then evening free

Monday

April 11th 2011

09.00 – 10.30

Lecture: A behavioural view of decision making

10.30 – 11.00

Coffee

11.00 – 12.30

Lecture: Multi-criteria decision making

12.30 – 13.30

Lunch

13.30 – 15.00

Group Exercise Session 1:  Introduction to the case study.

Groups meet and begin preliminary discussion on their case study facilitated by a course tutor

15.00 – 15.30

Tea

15.30 – 17.00

Computer lab exploring decision analytic software

17.00 – 18.00

Lecture: Decision support

18.00 –

Dinner

Tuesday

April 12th 2011

09.00 – 10.30

Lecture: Problem structuring methods

10.30 – 11.00

Coffee

11.00 – 12.30

Lecture: Handling uncertainty

12.30 – 13.30

Lunch

13.30 – 15.00

Group Exercise Session 2:  Groups continue working on their case study facilitated by a course tutor

15.00 – 15.30

Tea

15.30 – 16.30

Lecture: Developing consultancy skills: the decision analytic process

16.30 – 18.00

Exercise: Structuring problems and facilitating group discussions

18.00 –

Dinner and evening event

Wednesday

April 13th 2011

09.00 – 10.30

Group Exercise Session 3:  Groups finalise their analysis and prepare presentation working on their case study

10.30 – 11.00

Coffee

11.00 – 12.30

Group presentations on their case study and discussion

12.30 – 13.30

Lunch

13.30 – 15.30

Panel discussion: Publishing in top journals

15.30 – 16.00

Wash and brush up... how does the learning from the course fit with their doctoral studies

16.00 – 16.30

Tea

16.30

Departure

Facilities offered by Alliance Manchester Business School

As befits one of the world’s leading business schools, Alliance MBS offers a range of teaching environments.  We have a group decision support room equipped with a range of software, including GroupSystems’ ThinkTank, which will allow us to explore how decision support systems and software may be used in a variety of individual and group circumstances.  Teaching rooms are fully networked and equipped with computers, data projectors, etc.

Course tutors

The following have agreed to offer sessions:

  • Prof Simon French
  • Prof Jian-Bo Yang
  • Prof Theo Stewart
  • Prof John Keane
  • Dr Nadia Papamichail
  • Dr Ling Xu
  • Dr Ludi Mikhailov

We may also bring in one or two outside speakers including clients who have experienced decision analysis support to their decision making and can comment on its value.

Location

Manchester is a major UK city with extensive transport connections including an international airport.  The North West region has much history, particularly in relation to the Industrial Revolution.  There are many areas of outstanding natural beauty in easy reach: the Peak District, the Cheshire Plain, the Lake District and North Wales.  The city itself is vibrant with many attractions and entertainments.

Length of Course

The course will begin on the afternoon of Sunday 10th April and continue until the afternoon of Wednesday April 13th, 2011. There will be a social outing/event on the evening of the second day.

Accommodation

Accommodation is not included in the package offered to participants.   There are many hotels and hostels within walking distance of Alliance MBS and accommodation at these can be very cheap if various promotional offers are taken.  During the months of January and February such offers are available over the Internet.  Thus participants are advised to book early to take advantage of these.  We will provide a list of such hotels and hostels upon accepting the students onto the course.

There are also some funds available to offer some limited support to some, but not all of the participants. 

Applications and Registrations

Applicants should submit the attached form, preferably by email, along with a short c.v. to:

Professor Simon French

Alliance Manchester Business SchoolUniversity of ManchesterBooth Street WestManchester, M15 6PBUnited Kingdom

Tel: +44-161-275-6401

Email: simon.french@manchester.ac.uk

Places are limited, so early application is recommended. Do not book accommodation until we have accepted you on the course.  We will endeavour to turn around applications within five working days.  Acceptance will be conditional on our receiving the registration fee of £100.00 within 21 days of notification of acceptance or before April 1st 2011, whichever is the earlier.  Do not send the registration fee with the application.

Note that students are responsible for arranging their own visas and other travel documents and costs.  Upon acceptance we will issue a letter of invitation to the School such as may be needed by the UK visa authorities.

About

The Decision and Cognitive Science Research Centre has become a world leading centre of research excellence in the areas of Multiple Criteria Decision Analysis (MCDA) and Decision Support Systems (DSS).

Contact us

  • Address:
    Manchester Business School
    Booth Street East
    Manchester
    M13 9SS, UK

  • Phone:
    +44 (0) 161 820 8344