Expert systems can be used to solve difficult real world problems. An expert system can be thought of as a human expert inside a computer that will offer advice and guidance for a problem in their field of expertise. These systems typically have a knowledge base that contains information from one or many experts and an inference engine that takes user input and makes a conclusion based on the rules in the knowledge base.
In building science, it is difficult to predict the moisture durability of a given envelope component such as a wall or roof. The durability depends on all the materials used in the construction, the climate, and indoor conditions. Experts in the field have good intuition about how some common envelope designs perform. As building codes require more insulation and tighter construction, special care needs to be taken to ensure these energy efficient assemblies are moisture durable. To fill this knowledge gap, ORNL and the DOE are applying the rule-based expert system methodology to a web-tool that offers expert advice to builders and designers to help determine if a wall design will be moisture durable and provide guidance regarding successful implementation. The expert system is populated with knowledge from both experts in the field and probabilistic hygrothermal simulation results.