This paper presents a novel approach, which is based on integrated
(automatic/interactive) knowledge acquisition, to rapidly develop
knowledge-based systems. Linguistic rules compatible with heuristic
expert knowledge are used to construct the knowledge base. A fuzzy
inference mechanism is used to query the knowledge base for problem
solving. Compared with the traditional interview-based knowledge
acquisition, our approach is more flexible and requires a shorter
development cycle. The traditional approach requires several rounds of
interviews (both structured and unstructured). However, our method
involves an optional initial interview, followed by data collection,
automatic rule generation, and an optional final interview/rule
verification process. The effectiveness of our approach is demonstrated
through a benchmark case study and a real-life manufacturing
application.