The main purpose of this thesis is to develop an assistant system in the drug administration of anesthesia. The character of the system is like a virtual anesthesiologist. It gives advices to the anesthesia practitioner on drug administration during the procedure of anesthesia. The expectation is to maintain the anesthesia in the proper depth. Utilize the learning and the human-like reasoning ability of the fuzzy neural network. We designed the virtual anesthesiologist to accommodate the knowledge and the experience of the real anesthesiologist in anesthetic drug administration. We adopt the anesthesia simulator into the experiment environment to test the virtual anesthesiologist and to explore the value of the simulator. The experiment is divided into two different phases. The results reveal the preliminary capability of the virtual anesthesiologist.