Industrial Diagnosis, Planning and Simulation - Introduction
by Per Kreuger
The special theme of industrial diagnosis, planning and simulation is structured around a selection of industrial problems rather than a particular type of technology or methodology. We have chosen contributions describing a variety of techniques and applications that are of relevance to these problems, with the intent of highlighting state-of-the-art applied research within the ERCIM institutes and their associates.
Recently at SICS, the Industrial Applications (IAM) Laboratory was formed around a novel topic in applied research. We brought together a number of researchers with expertise in mathematics and computer science, who share the conviction that the study of the applicability of techniques and methods developed within the scientific community should itself be the subject of research.
The activities of such a group will, of course, always be influenced by the backgrounds of the people participating in it, not only with regard to their specific scientific expertise but also, and in this case significantly, by their familiarity with practical problems outside the scope of their scientific schooling. The group formed at SICS brings together researchers from the fields of combinatory optimisation, scheduling, flow theory, constraint programming, statistical methods for data analysis and learning systems. They have domain expertise in process and manufacturing industry and resource management, monitoring and optimisation in transportation and logistics.
The initiative behind the formation of this group came from the observation that methods and techniques from these scientific disciplines often have the potential to improve both the efficiency of resource usage and the quality of products and services, as well as reducing the costs associated with production. On the other hand, not all methods proposed by researchers scale well to the size and diversity of industrial problems.
The fact that techniques and methods developed in science are used to solve practical problems is, of course, nothing new. However, while this is the basis of all technology, the quality of the interplay between science and technology within a society determines to a large extent its potential for growth and development. In the case of computing science, which is the study of algorithms, their properties and use, one must always be careful to verify results against practical problems. This is a time-consuming and, at times, frustrating experience for many researchers, since their experience lies in a different field to that in which the problem occurs. Even so, this process is essential for the growth of technology and provides valuable feedback to the research discipline.
For example, within the field of optimisation and combinatorial reasoning there exist many elegant and highly applicable results for a wide selection of specific problems. In a few cases, an industrial problem will exactly match a certain problem idealisation and the mathematical model used to study the properties of a particular class of algorithms. This is, however, only rarely the case. Real practical problems in, for example, industrial production, are invariably complex combinations of several smaller problems and may have numerous related idiosyncratic conditions. The ability to understand and describe a practical problem in terms of the types of models and methods used in this field is usually referred to as modelling expertise. Interestingly, this expertise is rarely described in the scientific literature. The ability to use outside a scientific field the algorithmic methods developed within that field exists in parallel with the science, and is indeed rather more of an artform.
A systematic method for solving a class of practical problems is called a methodology. However, most methodologies used in computing science are based around a fairly limited set of algorithmic methods and quickly become useless whenever the problem changes too much or is combined with other problems into a more complex situation. This raises issues of method generality and scalability, which should be of prime concern to the scientific field developing the method as well as of great practical importance to the society in which the scientist works. A systematic study of these issues will lead to two types of results:
- better understanding of typical models for a selection of important real-life industrial problems
- better understanding of the properties of a selection of practically useful algorithmic methods.
Industrial applications of state-of-the-art techniques and methods in computing science allow researchers to test, in practice, the flexibility, scalability and utility of their techniques and methods. In addition, it is an opportunity to push the mature parts of the scientific field out into practical use for the benefit of the industry and society at large, be it in manufacturing, transportation, processing, telecommunications, biotechnology or the service industry. This is true for many fields of computer science, but particularly for those directly involved in the modelling and solving of typical problems occurring in industry, eg process simulation, monitoring and prediction, capacity analysis and allocation, fault detection and diagnosis, production planning, flow optimisation, resource scheduling and allocation, structure detection and matching. We believe there is a strong and fundamental need to systematically study the practical utility of such methods by applying them to a large number of real cases.
For the special theme we have aimed to highlight successful or promising applications of advanced techniques from these fields, including some in full-scale industrial settings, which we hope will contribute to our understanding of the strengths and weaknesses of the various techniques and methods. The articles have been grouped into four sections:
1. Electronics and Networks
2. Process Monitoring and Optimisation
3. Transportation and Logistics
4. Process Design and Management.
The first section contains two articles on the use of magnetic field models in the electronics industry, one on the management and optimisation of communications network resources, and a fourth on intrusion prevention.
The second - and largest - contains articles on monitoring, diagnosing and optimising production processes. Of these, several are concerned with the detection of deviations from normal process parameters, while others address issues in connection with identifying the cause of, and correcting, faulty behaviour. In a few cases, optimisation of process parameters and redundancy of monitoring systems are also relevant.
The third section contains articles on problems in transportation and logistics. This type of issue often contains instances of specific problems that are comparatively well understood, but in practice are frequently made very difficult to solve by combinations of sub-problems such as resource allocation, routing and scheduling. One of the articles describes an analytical method for capacity analysis and assessment using techniques from discrete event systems. Three are concerned with management and routing of vehicles, which is one of the most cost-intensive operations in transportation. Others describe
approaches to solving packing, placement and storage problems.
The articles in the final section describe methods for process design and project management - enjoy!
Per Kreuger, SICS