Process mining application for automatic auditing of HACCP log events in tuna companies

Abstract

Consumers are increasingly demanding with respect to the quality of the product they purchase, which at a business level means improving processes by adopting methodologies suitable for the management of information generated within the activities, which generates a large volume of data that is not easy to handle manually. Therefore, process mining takes a key role in the management of data in an optimal way, applied for automatic auditing of hazards and critical control points. This quality system, better known by its acronym HACCP, is a powerful tool to ensure product safety, hygiene and traceability. Normally, the analysis of the records is carried out at the auditor's discretion and by sampling due to the volume of data. For the management of this BIG DATA, there is process mining, a young discipline related to machine learning, data mining, modeling and process analysis. The present work aims to propose a method to apply process mining to improve the quality of automatic auditing of HACCP records in tuna companies. This research uses a qualitative methodology, with a survey of experts and research processes to collect information contributing to the importance of process mining in automatic auditing systems. A method is proposed to make process mining applicable for the automatic auditing of records on hazards and critical control points, which was validated with a case where all transactions were automatically analyzed.