Research

The unprecedented ease with which process data and information can be collected and recorded has motivated increasingly sophisticated process analysis, management and control and a general movement toward the deployment of “Smart Plant” technologies. Smart Plants are supported by intelligent systems that integrate data, information and knowledge to provide decision support for activities such as improving product quality and resource utilization, abnormal situation management, reducing product development cycles, facility reliability and HAZOP. 

Data, models, knowledge, and experience all have critical roles in management and decision support. Ever tightening operating requirements, increased demand for new products on shortened life cycles, increased business-to-business and business-to-customer interactions and the benefits of autonomous operation have led to the development of distributed intelligent systems that integrate data analysis and interpretation, information processing, and automated decision-making. 

Jim Davis' research program fundamentally considers how qualitative and quantitative information and knowledge are used. Studies are directed toward the development, application and integration of knowledge-based and neural reasoning techniques into decision support systems for process operations and design. Knowledge-based studies include data interpretation, knowledge representation, knowledge manipulation, and databases of engineering knowledge. Neural reasoning encompasses on-line pattern recognition, learning sets and adaptive pattern classification. There is particular emphasis on systems integration, considering both the integration of various knowledge-based and neural reasoning techniques with other information processing approaches and the integration of the resulting systems into the operating or design infrastructure of the process.

In general, chemical engineering graduate students both contribute to and take full advantage of multi­disciplinary views.  Student often work within a joint advisor relationship providing unique opportunity for deep involvement of multiple perspectives. Seminars, research presentations, exam committees, and group meetings often involve interdisciplinary participation. Students are exposed to a variety of viewpoints that often provide new ways of solving traditional chemical engineering problems or offer new opportunities for extending chemical engineering techniques.

To meet theoretical objectives, specific industrial applications are used to investigate, develop and/or progress new methodologies and approaches. Industrial problems are used to investigate and develop theories; the projects lead to specific theoretical and practical results; and graduate students gain an industrial perspective as part of their graduate work. The overall objective is a healthy interaction between academia and industry.

Technology integration represents a defining research emphasis within the research program. Projects focus on the development and integration of a variety of knowledge-based system techniques, neural reasoning approaches, conventional control methodologies, statistical approaches, filtering techniques, and numeric and symbolic simulations. Projects, however, have been defined in terms of functional objectives and not technologies. Examples of current projects include on-line diagnosis of continuous and batch operations, data interpretation, intelligent control, sensor validation and multipurpose knowledge repositories. The program objectives are being pursued through specific applications in the petroleum, chemical, paper, manufacturing, agricultural and electronics industries.