Emerald Group Publishing Limited
Copyright © 2011, Emerald Group Publishing Limited
Article Type: Guest editorial From: Journal of Manufacturing Technology Management, Volume 22, Issue 6
About the Guest Editors Khairy A.H. Kobbacy, PhD, MSc, BSc, is the Professor of Management Science, School of the Built Environment, University of Salford, UK. He has a long-standing interest in “applied” Operational Research. He previously lectured in Operational Research at Strathclyde University after gaining industrial experience as a Senior Petroleum Engineer in the Strategic and Production Planning Department of a major oil company. His research interests in operations management are directed towards the development of intelligent management systems for maintenance scheduling, inventory control and contract selection in supply chains. His research in maintenance has been funded by industry and the research councils. He published extensively in the area of maintenance management/modelling and is co-author of the Complex System Maintenance Handbook, Springer, 2008. He chaired four European conferences on Intelligent Management Systems in Operations, Salford in 1997, 2001, 2005 and 2009. Professor Kobbacy was awarded the Operational Research President’s medal in 1990 and the Literati Club Award in 2001. He has been Vice President of the Operational Research Society, UK.
Sunil Vadera, PhD, MSc, FBCS, CITP, C.Eng., is a Professor of Computer Science and the Associate Head Research for the School of Computing, Science and Engineering at the University of Salford. His research on AI has included use of machine learning for sensor validation with the Mexican Instituto de Electricas, the development of a system known as Dust-Expert that advises on the relief venting of explosions in chemical processes for the Health and Safety Executive, development of a Bayesian Exemplar model based on Wittgenstein’s family resemblance principle and research on data mining for the credit rating of sub-prime loans. He was Chair of the UK BCS Knowledge Discovery and Data Mining Symposium held in Salford in April 2009 and General Chair of the IFIP 2010 Conference on Intelligent Information Processing. Further details of his most recent research can be found on the web site: www.mendeley.com/profiles/sunil-vadera/
As business and industry become more global, diverse and market driven, the drive for more effective solutions for problems in operations management increases.
In recognition of this drive and the need to bring the artificial intelligence (AI) and operations management communities closer together, we established and held the first European Conference on Intelligent Management Systems in Operations (IMSIO) in 1997. This first conference attracted significant interest and was followed up with conferences in 2001 and 2005 that led to several special issues of journals including two issues of the International Journal of Manufacturing Technology Management (in 1999 and 2000). This special issue follows the fourth IMSIO Conference, held at Salford in 2009, and aims to highlight interesting and useful applications of AI in operations management and manufacturing.
At each IMSIO Conference, we have carried out a survey of the applications of AI in operations management, identifying trends, gaps and directions for future research. The first paper of this special issue presents a survey, that has grown from a relatively small study that was published in this journal in 2000 (Vol. 10 No. 4) to one where we have categorized over 2,600 papers from 1995 to 2009. The survey shows, for the first time, the trends in using AI methods in operations management over the last decade and gives an indication of future trends and directions of research.
In the next paper, Ken McNaught and Andy Chan present the use of Bayesian networks to develop a fault diagnosis system. The paper begins with the motivation for using Bayesian networks, includes a brief introduction to Bayesian networks, and presents an illustration of how they have used Bayesian networks for developing WISDOM, a system used by Motorola as part of their diagnosis and testing process for base transceiver stations. The paper includes an interesting comparison of the relative performance of the system when used by operators in China and the UK.
The paper by Anna Ławrynowicz is representative of a common direction of research, that of using genetic algorithms (GAs) for scheduling. The paper proposes modifications of GAs for scheduling in nodes of a supply chain and for scheduling in an industrial cluster. It includes a detailed comparison of the proposed modified GAs with some commonly used dispatching rules.
Mohamed Zaki, Babis Theodoulidis, and David Díaz Solís present a case study in the application of data mining for detecting fraud detection. The paper carries out a comparison of various methods including neural networks, decision-tree learning, and logistic regression for attempting to detect stock touting e-mails that aim to gain illegal profit by disseminating rumours and inaccurate information.
Khairy A.H. Kobbacy, Hexin Wang, and Wenbin Wang discuss the development of a prototype intelligent supply contract design system. The system utilises a rule based approach to help suppliers to choose the appropriate supply contract in order to maximise the supplier’s expected profit while not disadvantaging the buyer. The rule-base which is used to select the appropriate contract was developed from the results of an analysis and comparison of a number of a single- and two-period supply contracts that are frequently encountered taking into consideration relevant factors such as cost structure and demand pattern. Examples are given in the paper to demonstrate the effectiveness of the system.
The paper by Rashid Mehmood and Jie A. Lu presents a study of the computational cost of using Markov chains and queuing theory and proposes methods for reducing the computational cost of these methods, which is a known barrier to their use in practice. The paper includes a Markov model of a car-free city with some residential areas and medical centres to motivate the problems of increased dimensionality, describes some common numerical methods used to compute steady state solutions and presents methods for storing large Markov chains.
The final paper of the special issue, by Mohamad Saraee, Seyed Vahid Moosavi, and Shabnam Rezapour presents a practical application of self-organising maps (SOM) and decision-tree learning to model and control a multi-response machining process. SOM are used to identify and visualise the relationships between the cost effectiveness of the process, and the quality of a product. The application also illustrates how decision-tree learning can be used effectively to predict the quality of a product based on control parameters.
The special issue aims to encourage greater and more effective use of AI in operations management and we hope you find the papers interesting and useful.
The Guest Editors are grateful to the referees for reviewing the papers in this special issue.
Khairy A.H. Kobbacy, Sunil VaderaGuest Editors