This thesis describes the development of an integrated intelligent system for the operational support of a batch sulphite pulping process. The integrated intelligent system consists of several symbolic reasoning systems and numerical computation packages. Two kinds of software programs, which are developed to predict the final pulp quality by using both qualitative physics (symbolic reasoning systems) and a mathematical model (numerical computation packages), are written in different languages and used independently. All the programs are under the control of a supervising intelligent system, namely, a meta-system. The meta-system manages the selection, operation and communication of these programs.

The system, which is called "Intelligent Operation Support System" (IOSS) for batch pulping processes, provides assistance to human operators to achieve better operation of the batch pulping process. IOSS is the first prototype system developed for the pulping industry. IOSS is constructed using real industrial process data, so that it can be directly implemented in the plant. IOSS, which is a user friendly menu-driven system and a graphic environment, utilizes variable rules to organize its knowledge base. It also have a common sense reasoning mechanism to protect the knowledge from processing impossible physical information. IOSS, which can be run on any PC/MSDOS base machine, can also be employed for on-line supervisory control.