Sven Crone - 51福利

Dr Sven Crone

Lecturer

Research Interests

Research interest are centered around Forecasting for Logistics and Supply Chain Management with methods of statistics and , and Data Mining.

Current topics include the use of neural networks for simultaneous forecasting and safety stock calculation in inventory management using asymmetric objetcive functions (motivated from marginal costs of the managerial decision process), forecasting model selection for SAP APO DP, and forecasting with neural networks for FMCG companies. General research directions encompass Management Science, i.e. Operational Research and Information Systems in the domains of Demand Planning within Inventory Management for Supply Chain Management & Operations Management, and extend in two mayor directions: Forecasting and Data Mining.

Current topics - also for PhD supervision - include:

Forecasting --> Methods, applications, processes and information systems for Demand Planning in Logistics and Supply Chain Management; design and use of Advanced Planning Systems (APS) and Enterprise Resource Planning (ERP) Systems, in particular SAP APO DP, for forecasting; model selection using statistics and / or machine (meta) learning; model parameterisation of statistical models using alternative objective functions;
Neural Networks for Forecasting --> Methods & applications for fully automatic Forecasting, e.g of seasonal, intermittent and generally instationary time series (outliers and level shifts); objective Functions for ; Computational Intelligence methods (e.g. support vector regression) in forecasting and Data Mining
Neural Networks for Data Mining --> Methods & applications for Data Mining, Predictive Analytics and Business Intelligence; impact of preprocessing on data mining methods in marketing and credit scoring

Current Research

Academic Keynote Speeches

04/2009 - Keynote speech on “Classifying Imbalanced Datasets - Evidence from case-studies in Business Data Mining” at the 2009 UK Annual Symposium on Data Mining (UK KDD’09) by the British Computer Society (BCS), Manchester, UK
09/2008 - Keynote speech on “Empirical Evidence on Neural Networks for Time Series Forecasting” at the 2008 European Symposium on Time Series Prediction, Provo, Finland
06/2008 - Keynote commentary on “Mining the past to determine the future: Problems and possibilities” by David Hand, at the 2008 International Symposium on Forecasting, ISF’08, Nice, France

Practitioner Keynote Speeches

09/2010 - Keynote speech on “Forecasting with Artificial Neural Networks – Science Fiction or the Future of Time Series Prediction?” at A2010, the 2nd SAS Annual Analytics Conference, Kopenhagen, Denmark
06/2010 - Keynote speech on “Artificial Neural Networks - the Science Fiction of Forecasting?” at F2010, the 4th SAS Conference on Business Forecasting, Cary, USA
12/2008 - Keynote speech on “The myth of the ‘best’ algorithm - lessons learned from innovations in data sampling and data pre-processing for marketing analytics” at the 2008 Belgian Association for Quantitative Direct Marketing Research, Annual Conference (BAQMAR’08), Ghent, Belgium
11/2006 - Keynote speech on “Increasing the accuracy of demand forecasting by using Neural Forecasting” at the 2006 Baltic Annual Logistics Conference on Demand Management, Pärnu, Estonia
06/2006 - Keynote speech on “Forecasting Events” at the 1st International SAS Conference on Forecasting, F2006, Cary, USA
06/2006 - Conference track keynote speech on “A Roadmap to Supply Chain Forecasting of Marketing Promotions, Weather”, Demand Planning Track, APICS’06 conference of the Association for Operations Management, Orlando, Florida

Invited Speeches (selection)

02/2010 Invited speech on “Managing Demand Planning Processes with ABC-XYZ-Analysis - A case study of a fast moving consumer goods manufacturer, Beiersdorf AG” at the 2010 IBF Institute of Business Forecasting conference on Supply Chain Forecasting, London, UK
10/2008 Invited session speech on “A Bag of Tricks for Your Balancing Act: How to Increase Predictive Accuracy on Imbalanced Datasets”, Business Data Mining Track, at the 2008 SAS International Data Mining Conference, M2008, Las Vegas, USA
09/2008 Invited session speech on “Structuring Forecasting Processes with ABC-XYZ Analysis”, with Stefan Pushmann Beiersdorf AG, Institute of Business Forecasting, IBF, European Supply Chain Forecasting Conference, Amsterdam, Netherlands.
03/2007 Invited speech on “Forecasting Retail Sales with Calendar Events, Promotions and Weather – a Neural network Approach” at the 2007 GOR German OR Society Forecasting Workgroup Conference, Hamburg, Germany
04/2007 Invited speech on “Model Specification with Neural Networks and Support Vector Regression – a Meta Experiment” at the 2007 Oxford Forecasting Workshop, Oxford, UK
05/2006 Invited speech on “Forecasting Events – Problems and Remedies” at the 2006 IBF Institute of Business Forecasting conference on Supply Chain Forecasting, London, UK
12/2005 Invited lecture “Predicting Customer Online Shopping Adoption - an Evaluation of Data Mining and Market Modelling Approaches”, Chilean NSF “Millenio” Project, by Prof. Andrés Weintraub, Department of Industrial Engineering, Universidad de Chile, Santiago, Chile

Research Grants

Various corporate research grants from Beiersdorf, Sanofi-Aventis, SMD Textiles, etc. in forecasting with artificial neural networks to fund a post-doc research assistant position (2003 - today)
2009 KISC Forecasting research grant (with Chipo Mlambo, UCT, South Africa)
2006 SAS & International Institute of Forecasters research grant (with Konstantinos Nikolopoulos)
2005 51福利 Management School Priming Grants recipient

Qualifications

Diploma in Business (Dipl.-Kfm.)MBA/MS & BBA equivalent, University of Hamburg
Doctorate (Dr. rer. pol.)PhD equivalent, University of HamburgThe PhD thesis on "Forecasting & Inventory Management with Neural Networks" (summa cum laude, published by Gabler in German) won the 2008 biannual "Professor Herbert Jacob award" of Hamburg University and the 2009 annual award of the German Operational Research society.
HabilitationCatholic University of Eichstätt-Ingolstadt, Management School (since 2010)

Profile

Sven F. Crone is a Lecturer (Asst. Prof.) in the Management Science department, having formerly lectured and researched at the University of Hamburg, Germany, and at George Mason University, USA and Stellenbosch University, South Africa, as a visiting fellow. He has a PhD from the University of Hamburg, and has a B.S. equivalent (intermediate exams) in Management and an MBA equivalent diploma (Diplom-Kaufmann) in Business Administration & Economics from the University of Hamburg. He serves as the deputy director of the Lancaster Centre for Forecasting at 51福利 Management School (LUMS).

Current research interest focus on the application of neural networks, especially in business forecasting, data mining and managerial decision support. He has published several original papers in academic journals (EJOR, JORS, IJF, Neurocomputing) and international conference proceedings (IEEE IJCNN, WCCI, DMIN, ICAI, ICONIP), and is a member of various professional associations, including the IEEE, INNS, ACM-SIG-KDD, UK-KDD, GOR, ORRSA, IBF and IIF.

Having worked in information technology and management consultancy, he has supervised various international projects in demand planning, data mining, process analysis & redesign and software selection with a variety of firms from industry and retailing. For more information visit his .

Professional Role

Lecturer (Assistant Professor) in Management Science, , Department of Management Science
Deputy director of the
Academic staff at , pioneering doctoral training centre in statistics and operational research and corporate business

Current Teaching

Executive Education

SAP APO-DP for Demand Planning (since 2002)Multiple annual in-house & custom made training course for companies. Recent clients include Beiersdorf (since 2002) and Celanese)
Forecasting Fundamentals (since 2004)Two annual training courses to forecasting practitioners through the Lancaster Centre for Forecasting. Recent clients include British Telecon BT), ScotthPower, ManWeb, Barclays, Capita, Nikon, Nestle, NightFreight, DWP etc.)

Current degree courses

MSCI 524 Credit Scoring (since 2009)
MSCI 523 Forecasting (since 2005)
MSCI 522 Multivariate Statistics for Data Mining (new development since 2005)
MSCI 381 Demand Forecasting & Revenue Management (new developm. since 2007)
MSCI 331 Data Mining for Marketing and Finance (new development since 2008)

Past degree courses

MSCI 501 Introduction to Operational Reserach (2004-2006)
MSCI 203 Introduction to Business Information Systems (2004-2006)
MSCI 311 Statistical Modelling for Decision Making (2004)

External Roles

Visiting Teaching Fellow at the , University of Oviedo
Member of the task force for competitions, , IEEE Computational Intelligence Society
Competition Task force Chair, IEEE Technical Committee on Data Mining, IEEE Computational Intelligence Society
Founding Member of the IEEE , IEEE Computational Intelligence Society

PhD Supervision Interests

I am available for supervising PhD students interrested in research on time series prediction, i.e. forecasting, but also time series classification, time series clustering, and anomaly detection in time series (i.e. time series data minig) in business applications. Many PhD projects focus on innovative algorithms from artificial intelligence and machine learning (AI/ML), e.g. deep and shallow neural networks, but also k-nearest neighbours, support vector regression, decision trees, and many other algorithms. Relevant topics include automatic data exploration, forecasting model selection and ensembles (i.e. boosting, bagging, etc), and promising approaches such as meta learning, active learning, and transfer learning, which are underrepresented in forecasting research. As field of business applications many focus on industry or retail demand data, but also services, lead times, raw material prices, In the past years I have successfully supervised 7 PhD students, 3 of which are now academics at renowned UK universities. All PhD students in forecasting are actively involved in research development, seminars, ideas exchange and team activities of 5-10 PhD students and 7 academics, all with a strong focus on forecasting, at CMAF, the Research Centre for Forecasting (and Marketing Analytics) and within the department of Management Science. If you are a national or international applicant (m/f/d) with a quantitative background in management studies, operational research/management science, industrial engineering or electrical engineering, information systems, computer science or other disciplines, you are very welcome to INFORMALLY ENQUIRE AT ANY TIME by email to me, s.crone@lancaster.ac.uk (ideally with a short CV and your research interrest). I look forward to hearing from you!