Application of Natural Language Processing in Procedure Design

Abstract

Procedures play a vital role in high-risk systems in ensuring that the required level of system performance is achieved while maintaining the associated risks below an acceptable level. An effective procedure has a low level of task complexities i.e., elements contributing to a multiplicity of paths/outcomes, uncertainty in information, interdependency, a multiplicity of instructions/object of instructions and an excess amount of information. As a part of this study, we have developed a Natural Language Processing algorithm that can identify these elements in a procedure. The algorithm was tested for a dataset consisting of 20 procedures and the results were found to be promising.