Objectives: To trace software requirements and remove incomplete and inconsistent requirements of software and know the priority of requirements in a garment production management system. Methods/Statistical analysis: In multi-objective teacher-learning artificial bee colony optimization technique, optimal requirements are selected with the consideration of multiple constraints and multiple objectives. The main aim of this technique is to reduce cost and time consumption and maximize the client satisfaction that make use of constraints like cost threshold constraint, time threshold constraint and interaction constraints. Thus the selected requirements are traced for the purpose of removing the incomplete and inconsistent requirements. Requirement tracing is a process of requirement management that traces the requirements through their life cycle. The requirements are traced using TD-IDF weighting schema and firefly algorithm which are utilized to trace document requirement to requirements. From the traced requirements, there may be incomplete and inconsistent requirements that is removed by using logic based representation of requirements. It improves the quality of the software. Finally, complete requirements of the software is obtained and then prioritize the requirements based on case based ranking method to know which requirement is more important than the other. It follows three step process which are pair sampling, priority elicitation and priority learning. Priority learning is enhanced by semi- supervised ranking with graph based regularization where the semi-supervised labels are learned about the priority. Findings: From the selected requirements of software, trace those requirements from the document requirement to requirements and reject the requirements contain incompleteness and inconsistence and ranking the requirements to know the priorities of requirements. Improvements/Applications: The experimental results prove that the proposed techniques perform better than the existing technique in terms of precision-recall curve, threshold-f measure curve, precision and recall.
