Expert systems applications, computing environment

Until recently, most expert system building took place in the research departments of universities and a few major corporations. The primary emphasis was investigation of artificial intelligence principles, and the application was of secondary importance. The expert systems tools used reflect this interest. They are typically stand-alone AI computer systems, using special hardware and software environments (usually Lispr-based) not commonly fo md in scientific and engineering organizations. 

Technology is one of many trends constantly affecting computer validation. One area requiring special attention is the introduction of expert systems into the FDA-regulated environment. The simplest form of artificial intelligence generally used in applications (such as mortgages, credit card authorization, fraud detection, e-commerce, personalization) is the rule-based system, also known as the expert system (ES). 

In manual calculations, such problems could easily be spotted by an independent, critical observer, as one of the points made in the Study Director s statement shown in figure 34 (see page 271) will demonstrate. In an electronic system only an expert programmer, however, could be expected to deal with such issues, and furthermore only so, if he had full access to the source code. Therefore, the basic process for ensuring GLP compliance of an IT application is to demonstrate that the respective application does indeed perform the intended task in a correct way, and that it does so in the actual working environment of computer hardware and connected peripheral components and apparatus. This demonstration of the suitability for its intended purpose is obviously of fundamental importance, and this process is referred to as computer validation . In essence, such validation should provide a high degree of assurance that the system will meet its predetermined specifications. 

---