Although automation has become widespread in many industries, some workplaces, such as utility industries, still rely heavily on individuals to perform critical tasks based on their extensive past experiences. This accumulated intellectual capital not only improves productivity and efficacy under normal conditions but more importantly, also contains knowledge about identifying anomalies and addressing unexpected events and situations. This knowledge is essential and critical to ensuring system safety and reliability, especially for industries with aging infrastructures where anomalies are becoming more common. Unfortunately, this knowledge is not explicitly recorded in defined guidelines, protocols, or standard workflows but implicitly resides in the minds of skilled workers and routine event logs. As skilled workers retire or leave the business, we may lose this accrued indispensable knowledge that is critical to the industries’ productivity, reliability, and safety. This thesis proposes a framework to discover implicit knowledge from event logs. The implementation of the framework contains three phases. We first proposed an approach that uses hidden Markov models to filter out outliers from event logs in the clean phase. Then, in the discovery phase, we propose approaches to discover the implicit knowledge related to timing constraints and process scenarios from event logs. For timing constraint discovery, we have presented an approach that extends existing process mining techniques to mine and integrate timing constraints with a workflow or process model constructed by any existing process mining algorithm. A real-life road traffic fine management process scenario is used as a case study to investigate the effectiveness and validity of the approach. For process scenario discovery, we present the distance-based and density-based approaches that obtain timing information from event logs and use the information to assist process scenario discoveries. A wastewater treatment process provided by a domain expert is used as a case study to investigate the effectiveness and validity of the approach. In the incorporating phase, we present an approach that incorporates domain knowledge to assist in discovering process scenarios from event logs. The case studies and experiment results indicate that the proposed implicit knowledge discovery framework can mine implicit knowledge related to the timing constraints and process scenarios from event logs and outperform commonly used approaches with different real-life event logs.