A fire alarm system has numerous devices that work together to detect and give warning to the people through visual and audio appliances when there are smoke, fire and gas. The system is very sensitive to the fire, smoke and gas; hence the system sensitivity must be advanced enough so that it does not trigger any false alarm. The aim of this study is to reduce false alarm within the fire alarm system that diverts emergency responders away from legitimate emergencies that could result in loss of life and properties. The method used to conduct the research is a fuzzy logic approach (FLA). The method is tested using MATLAB and it has 125 rules since it has three variables which are fire, smoke and gas with five linguistic variables. The number of false alarms can be reduced if the fuzzy logic approach is put into practice in the alarm system since the probability of occurrence shows only 3% of error which is considered to be small. From these findings, we found that the number of false alarms can be minimized to the minimal by implementing fuzzy rules into the alarm system.