An expert system is a software representation of a specific domain of knowledge that can perform at or near the level of human experts. For example, in filter backwashing, an expert system would incorporate factual and heuristic data based on human plant operating experience, manuals, and plant physical and chemical relationships. This paper presents a case study of the City of Detroit Water and Sewerage Department's development of an expert system for filter backwashing, from listing what the expert system should be able to do through field testing and lessons learned.