Abstract
Roof maintenance is an issue that has ongoing ramifications for property owners, all over the world. Determining accurately what maintenance should be taken is difficult, with often-conflicting evidence further complicating the task.
A system has been developed for use by Roof Maintenance Experts. The expert can input information about the condition of the roof then a Fuzzy Neural Network makes an assessment, returning probable roof maintenance options.
This is a non-trivial problem from the realworld domain. Often, many combinations of possible maintenance can occur and, as individual parameters change, so does the prime (or most important) option.
A fuzzy neural network system was developed for assessment, running on a handheld device that could be taken into the field by a roof maintenance expert.