Alarm management is a fast-growing and important aspect in the petroleum operation domain. Alarm devices have become very cheap leading petroleum equipment manufacturers to overuse them transferring safety responsibility to operators. Not rarely, accident reports cite poor operators understanding of the actual plant status due to too many active alarms. Typical alarms for a process plant could average over fourteen thousand per day. Therefore, it is mandatory to have a ﬁltering process to distinguish expected from non-expected behavior during emergency scenarios. Ambient Intelligence contributes by enriching the petroleum plant environment with technology (mainly sensors and devices interconnected through a network) and built a system to help plant operators to make decisions based on real-time information gathered and historical data accumulated. Ambient Intelligence puts together all these resources to provide ﬂexible and intelligent services to users acting in their environment. Inspired by the distributed and encapsulated aspect of the process plant artifact physical model, we proposed a multi-agent-based alarm management system to analyze the process plant situation during emergency situations and assisting plant users to make sense of alarm avalanche scenarios.