(530c) An Ontological Knowledge-Based System for Identification of Efficient Chemical Production Routes

In process engineering, Knowledge-based systems (KBSs) have been used extensively to manage complexity, share information and (re)use information and data to solve problems related to, for example, the selection of process monitoring and analysis tools (Singh et al., 2010), enterprise performance measurement (Wen et al., 2008), design of biotransformation process (Dervakos et al., 1995) to name a few. There is a wide range of problems where KBSs have proved to be valuable tools. Ontologies have been used extensively by information technologists to systematically represent the knowledge within a given domain and to transfer and reuse knowledge across businesses. However, much less attention has been paid to the design and development of a knowledge-based system that can be shared among the users from the same organization or different organizations and can be applied systematically for the identification of appropriate production routes of a given chemical.

In most of the manufacturing industries such as chemical and pharmaceutical industries, it is highly desirable to identify the most optimal and efficient production routes for a given chemical products (e. g. production of an API). For a targeted existing/new product, several production routes can exist. However, the identification of an efficient chemical production route for a chemical is still a challenging and time intensive task because of the different levels of complexity that are involved in the identification process. Therefore, systematic methods and tools are needed through which a suitable chemical production route for any product can be identified in the most efficient way.

In the work reported here, an ontological knowledge-based system, which is part of a method to determine the efficient production routes for specific chemical products, has been developed. The identification of an efficient chemical production route involves the selection of suitable reactants and reagents, identification of efficient reactions for transforming the reactants into the desired product, identification of suitable solvents for substances involved in the reactions, identification of suitable catalyst and the identification of the unit operations involved and the corresponding operating conditions. All these data (knowledge) need to be managed during the solution of a specific synthesis problem. Furthermore, during solution, additional data is generated through appropriate models. The management of the static and dynamic data together with an inference system to retrieve the knowledge/data constitutes the ontology-based knowledge management system.

According to the architecture of the knowledge base there is a main section of the knowledge base which is connected with the different more specific sections of the knowledge base. The different classes of the main section of the knowledge base are: products, reactants, reactions, reagents, catalysts, and solvents. Specific sections of the knowledge consist of specific knowledge/data domains, including properties of substances (reactants, products, reagents, solvents, and catalysts), reaction characteristics and the characteristics of the unit operations. Each specific section (e.g. reactants properties section) also consists of different classes (e.g. quantitative properties, qualitative properties), subclasses (e.g. fixed quantitative properties, dependent quantitative properties) and corresponding instances (e.g. density).

The objective of this presentation is two-fold: first to highlight the developed ontological knowledge-based system and second to demonstrate its applications through a case study involving the production of active pharmaceutical ingredients (API).

References

1. Singh, R., Gernaey, K. V., Gani, R., (2010). An ontological knowledge based system for selection of process monitoring and analysis tools. Computers & Chemical Engineering, doi:10.1016/j.compchemeng.2010.04.011

2. Wen, W., Chen, Y. H., & Chen, I. C., (2008). A knowledge-based decision support system for measuring enterprise performance. Knowledge-Based Systems, 21, 148?163.

3. Dervakos, G. A., Woodley, J. M., Washbrook, J., & Lilly, M. D., (1995). DESIGN OF BIOTRANSFORMATION PROCESSES, Use of a Knowledge-Based System. Trans IChemE, 73, 133-195