This thesis presents a novel technique — referred to as trend-weighted rule-based expert system (TWRBES) — grounded in the integration of two existing tools of the artificial intelligence field, expert systems (ES) and qualitative trend analysis (QTA). Main goal of this approach is to benefit of the major advantages associated with each of the techniques used, such as the ability to represent knowledge through rules and the capability to extract the behaviour and the trends of a continuous signal. The proposed methodology fills a gap between purely quantitative and purely qualitative methods, allowing to achieve results based on both the exact values and the trends of a given signal. Thus, the discussed technique allows the extraction of a certainty factor regarding a rule previously developed by an expert, ruling out the true/false logic used in classic expert systems. Such integration allows a direct purpose in industrial environment applications, especially in the intelligent automation field. The features of the proposed algorithm, particularly in terms of industrial process monitoring, are supported by simulations and experimental results based on industrial benchmark known as Tennessee Eastman Process.