CONTENTS

In the process industries, the biggest challenge facing process engineers will be the reduction of variable costs whilst maintaining product quality. Advanced process control is the most effective technology available to realise this objective, especially on established plants. As systems become more complex, another important aspect is the reliability of the implemented systems. Here, the reliability of hardware and software are issues which have to be addressed. Allied to this is the requirement for suitably designed man-machine interfaces to enable efficient and reliable information transfer and to facilitate systems management.

With regard to the primary modules making up an advanced control project, neural networks, nonlinear systems theory, robust control, knowledge based systems are areas which appear to have captured the attention of both researchers and practitioners in the field of control engineering. This trend will continue well into the next decade. Areas that will receive particular attention will be techniques that will translate raw data into useful information; improved measurement methods including inferential estimation; multivariable non-linear predictive control and formal techniques for analysing the integrity of neural network based methodologies.

All information is of value, and should not be discarded just because they do not conform to a particular model building procedure. Thus, new modelling methods are also required. These should provide a framework where a priori knowledge of the process could be combined with the various existing modelling techniques, leading to so called 'grey-box' models. The resulting models should also be amenable for utilisation by the different modern controller designs, thus rendering controller synthesis independent of model types.

The process industries have an enormous base of manufacturing facilities which are still being run by unsophisticated or primitive control schemes. Competitive pressures will not allow any company in these industries to ignore the significant efficiencies possible through adopting modern process control technologies. A major obstacle to realising the full potential of modern control techniques is the lack of exposure to advances in the field. This can be overcome by the development of portable computer based training packages. The current proliferation in multi-media computing systems is the ideal impetus for the development of such learning aids.