Development of an expert system

The appearance of expert systems to solve practical problems, also in chemistry, started in the eighties. During this period much experience has been acquired through the expected and unexpected problems that arose during such projects. Until now there are only a few commercially available expert systems and this is not likely to change in the near future. This implies that expert systems will be mostly in-house developments. The different steps to consider are  

---

R. Banares-Alcantara, A. W. Westerberg, and M. D. Rychener, Development of an Expert System for Physical Property Predictions, Technical Keport, Carnegie Mellon University, Design Research Center and Robotics Institute, Pittsburgh, Pa., 1983. 

Utilization of intelligent systems in chiral chromatography starts with an original project called CHIRULE developed by Stauffer and Dessy [36], who combined similarity searching and an expert system application for CSP prediction. This issue has recently been reconsidered by Bryant and co-workers with the first development of an expert system for the choice of Pirkle-type CSPs [37]. 

Barratt MD, Castell JV, Miranda MA, Langowski JJ. Development of an expert system rulebase for the prospective identification of photoallergens. Photochem Photobiol B Biol 2000 58 54-61. 

Thus, propanol, C3H70H, has a membership of 1 in the three-carbon molecule class, while ethanol, C2H5OH, has a membership of 0 in the same class. As the membership in a crisp set must take one of only two possible values, Boolean (two-valued) logic can be used to manipulate crisp sets. If all the knowledge that we have can be described by placing objects in sets that are separated by crisp divisions, the sort of rule-based approach to the development of an expert system described in the previous chapter is appropriate. 

The application of thermomorphic solvent systems as a new recycUng concept was investigated in various C - C bond-forming reactions. Therefore methods for a systematic choice of solvent combinations were developed. In addition to common organic solvents more unusual solvents Hke cycHc carbonates, pyrroUdones, polyethylene glycols and lactones were used in the investigations. The phase behaviour of the new solvent systems was determined by cloud titrations. From these experiments information about the temperature dependency and an appropriate composition for the reactions could be obtained. The results were used in the development of an expert system for the solvent selection. 

Betteridge, D., Mackison, R., Mottershead, C.M., Taylor, A.F. and Wade, A.P. (1988) Development of an expert system for the selection of sample points for moisture analysis. Anal. Chem., 60, 1534-1539. 

Ames BN, Durston WE, Yamasaki E, Lee FD (1973) Carcinogens are mutagens a simple test system combining liver homogenates for activation and bacteria for detection. Proc Natl Acad Sci USA 70 2281-2285 Barratt MD, Castell JV, Miranda MA, Langowski JJ (2000) Development of an expert system rulebase for the prospective identification of photoallergens. J Photochem Photobiol B 58 54-61... 

The aminoalkyl and the cyanopropyl bonded phases are commonly used in that they can be equally well used in either the normal phase or the reversed phase mode. The versatility of the cyanopropyl column was exploited in the development of an expert system for HPLC, allowing a single column to be used for all solutes. 

Barratt MD, Basketter DA, Chamberlain M, Payne MP, Admans GD, Langowski JJ (1994) Development of an expert system rulebase for identifying contact allergens. Toxicol In Vitro 8(4) 837-839... 

Resource estimation is a critical factor influencing the success of expert system projects. It is important that the developer accurately estimates the time, staff, and financial resources required to complete the project. Another resource estimation factor is allowing for experimentation or exploration through prototyping, often both are necessary in the development of an expert system. 

Homsher, M. T. Olivero, R. A. Robertson, G. L. Moore, J. M. Fisk, J. F., Development of an Expert System for the Analysis of Laboratory Performance Evaluation Data, In Proc. 3rd Annual Lockheed Artificial Intelligence and Strategic Computing Svmp.. Houston, Texas, 1987. 

Based on the proven track record and widespread applicability of the CPE/CCP approach, EPA s Office of Research and Development/Center for Environmental Research Information decided in 1988 to fund development of an expert system to incorporate the expertise embodied in the CPE/CCP methodology. This report describes the concept, design, and implementation of this expert system. The system, POTW EXPERT, was beta-tested in September, 1989, and is currently in limited-distribution testing by EPA. 

The Activated Sludge Advisor Prototype (ASAP) was designed to help the Dallas Central Wastewater Plant s novice operators control the activated sludge process. Our ultimate goal is to develop an expert system which deals with all aspects of plant functioning. The activated sludge process was chosen as the focus of our initial effort because (a) it is crucial to treatment plant functioning, (b) it entails a considerable amount of expertise, and (c) its circumscribed nature would permit development of an expert system within a reasonable time frame. 

In spite of the many perceived benefits, the development of expert systems per se over recent years has been surprisingly slower than one would expect. One possible explanation for this is that when the systems were first introduced, their capabilities were overestimated and they were seen as the panacea to all formulation problems. This was obviously not the case, but as a result, the systems are viewed with some degree of scepticism. Several reviews on the issues and limitations with the development of an expert system have been published (Dewar 1989 Tinsley 1992 Rees 1996). 

The development of an expert system requires a understanding of the functioning and thinking patterns of a human expert. Such an understanding often eludes even that expert. To be an expert, a person must be so familiar with a knowledge area that predictions or solutions to problems seem to flow from that person with little visible effort. This is in fact a characteristic of college... 

This project is unique in that it is one of the first attempts to develop an expert system shell specifically designed for chemistry. While other researchers have implemented expert systems, all have either developed their own shell or utilized a commercially available shell. CHESS will assist in the development of an expert system in the chemistry arena, allowing researchers to focus more on the problem being solved instead of computer science oriented details. 

Jung, E. S., and Freivalds, A. (1990), Development of an Expert System for Designing Workplaces in Manual Materials Handling Jobs, in Computer-Aided Ergonomics, W. Karwowski, A. Genaidy, and S. S. Asfour, Eds., Taylor Francis, London, pp. 279-298. 

There are many different opinions as to when a set of programs and data can be considered to be an expert system. A computer scientist working on the development of an expert system would favor the following definition ... 

For a time there was a lot of interest in expert systems. The concept makes good sense. Development of an expert system was not always easy to accomplish. Some were not effective. Expert systems, both inductive and deductive, may prove to be useful computer tools. They help perform some reasoning. 

Theories of instructional design do not suffice they must be supplemented with knowledge from other sources, such as studies concerning the implementation and evaluation of instruction. The knowledge of human experts, including experienced teachers, must be tapped. Consider the development of an expert system to do oil-well analysis although there exist models upon which the system can be based, the models are not sufficiently detailed to map out all situations, possible variations, etc. Input from human experts is also required. The task of developing an instructional system is no less complicated. Instruction is a complex domain, in which it is difficult to determine the relevant factors, let alone assess them. 

Baiiares-Alcantara, R., Westerberg, A.W. and Rychener, M.D. 1985, Development of an expert system for physical property predictions. Comp. Chem. Engng. 9,127. 

According to O Keefe and Preece (1996), the process of knowledge acquisition is the most difficult challenge for the development of an expert system. The quality of knowledge acquired supports on the quality of the analysis results obtained of expert system. 

The proposed method is suitable for acquiring knowledge for the development of an expert system applied in the diagnosis of deviations in the process including operational failures in order to search for the components which fail can cause these deviations. It is emphasized the importance of following the proposed flow chart that shows the sequence of activities that allows to make detailed and reliable analysis, because they are used and organized in a logical order, so that the method is applied correctly. 

Wanklyn, J. N., and Wilkins, N. J. M., Development of an Expert System for Design Consultation on Marine Corrosion, British Corrosion Journal, 20 161-166 (1985). 

---