tel: +33-4-92948493
Abstract
One of the essential requirements on the operation of complex industrial processes is to deal with uncommon and unusual situations. The need for the continual re-configuration of plant and the constant modification of operating conditions to meet production demands, means that operating staff are required to face situations where they cannot react or decide on the basis of their experience alone. The large number of possible situations makes it very difficult, if not impossible, to enumerate the correct responses for use in a decision tree or expert system based approaches. One solution to this problem is to utilise simulations of the process to train operators to recognise such situations and to investigate the effectiveness of possible control decisions. However, simulators that industrial processes are presently provided with are intended for design purposes rather than for operations guidance and are therefore too complex and detailed for operating staff to understand and use effectively. Further, this lack of usability makes it difficult to use simulations for on-line decision support, thereby limiting operations to known a priori situations. The industrial need is for simulation systems that allow direct interaction with the user and consequently are much easier for operators to use. Such Intelligent Simulation Systems can then be used for effective on-line decision support and training situations, thereby improving the efficiency, safety and quality of the associated industrial processes.