Ingenieur Vol. 74 Ingenieur Vol 72, April-June 2018 - Page 74

INGENIEUR Table 1: Differences between Conventional Systems and Expert Systems (Negnevitsky, 2005) Conventional Systems Expert Systems Information and processing combined in a single sequential programme The knowledge base is separated from the inference engine The programme is never wrong The programme could make a mistake. Complete input data is required. Incomplete input data is acceptable. Changes of programme are time consuming. Changes of programme can be made with ease. Quantitative data Qualitative data Numerical data representation Symbolic data representation. level problem-solving, expert systems generally: (Negnevitsky, 2005). 1. Support inspection of their reasoning processes, both in presenting intermediate steps and in answering questions about the solution process. 2. Simplify and ease modification which enables adding and deleting skills from the knowledge base. 3. Reason heuristically, using knowledge to obtain useful problem solutions. The reasoning of an expert system is open to inspection to provide useful information about the state of art and to explain the choices and decisions made by the programme. Explanations are important in expert systems for many reasons. Doctors or engineers who seek recommendations from an expert system must be satisfied with the solution. The explanations help people relate the advice with their existing understanding of the domain and apply it in a more confident and flexible manner. The major application area of an expert system includes (but is not limited to) the following: interpretation, prediction, diagnosis, planning design control and instruction. (Shu, 2005). Table 1 shows the differences between conventional systems and expert systems. Expert systems are being applied to a variety of problem domains to assist in the decision making 6 72 VOL 2018 VOL 74 55 APRIL-JUNE JUNE 2013 process (Chen et al., 2012). It is able to provide solutions for different problems in the industry ranging from planning to marketing strategies (Li, 2005) to re-engineering consultation (Hvam et al., 2004; Ruiz et al., 2011). Nowadays, expert systems are applied in various fields with different applications, such as: ● ● Constructing a nutrition diagnosis expert system (Chen et al., 2012) ● ● An expert system for diagnosing problems in boiler operations (Artificial et al., 1992) ● ● An expert system hybrid architecture to support experiment management (Fiannaca et al., 2014) ● ● A knowledge-based expert system for selection of appropriate structural systems for large spans (Golabchi, 2008) ● ● Expert system for making durable concrete for aggressive carbon-dioxide exposure (Islam & Miah, 2011) ● ● Expert system for Sri Lankan solid waste composting (Jayawardhana et al., 2003) ● ● A prototype rule-based expert system for travel demand management (Mansyur et al., 2013) ● ● Development of an expert system for the design of airborne equipment (Kumar et al., 2004) ● ● An expert system development tool for non- AI experts (Ruiz et al., 2011)