The advance of distributed control systems in the chemical industry has made it possible to install many alarms cheaply and easily. While most alarms help operators detect an abnormality and identify its cause, some are unnecessary. A poor alarm system might cause alarm floods and nuisance alarms, which reduces the ability of operators to cope with plant abnormalities because critical alarms are buried under many unnecessary ones. Typical nuisance alarms are sequential alarms, which are a collection of many alarms that almost always occur simultaneously with specific time lags within a short time. We propose an analysis method of similarities between sequential alarms by using the Levenshtein distance, which is a string metric for measuring the difference between two sequences defined as the minimum number of edit costs needed to transform a string into another with edit operations of insertion, deletion, or substitution of a single character. The proposed method first converts the plant operation data, which consists of occurrence times and tag names of alarms and operations, into sequential alarm data. Then, similarities between all combinations of any two sequential alarms in the plant operation data are evaluated using the Levenshtein distance. The number of sequential alarms is effectively reduced by grouping them in accordance with the degree of similarity. The proposed method was applied to simulation data of an azeotropic distillation column. Results showed our method is able to group a lot of sequential alarms into small number of groups in accordance with the degree of similarity.