Data-Driven Optimization of the Defrosting Process in Bolton Food through Design of Experiment (DOE)

This thesis examines the development of a systematic approach to optimize the tuna defrosting process at Bolton Food, a critical phase in the production workflow that significantly impacts product quality, operational efficiency, and sustainability. Leveraging the Design of Experiments (DOE) methodology, this study focuses on mitigating process inefficiencies, temperature variability, and supplier-driven differences in loin geometry, aiming to establish a robust framework for process control and improvement. The research begins by mapping the current defrosting workflow through on-site observations and consultations with personnel, revealing operational constraints and sources of variability. This foundation is reinforced with data-driven analyses that identify the most influential factors, such as water temperature, spray duration, and supplier-specific characteristics. Two experimental designs4full factorial and space- filling4are proposed to address these variables. To ensure reliable results, a new sensor configuration is introduced, featuring advanced temperature probes and data loggers, supported by a comprehensive experimental protocol. This study achieves several milestones: the identification of process inefficiencies, the development of a DOE tailored to Bolton Food's operations, the implementation of innovative monitoring systems, and the design of a detailed experimental protocol. Upon completing the DOE, the optimized defrosting configuration is expected to significantly enhance production yield and product consistency, supporting Bolton Food's strategic goals of sustainability and quality. Furthermore, the methodology outlined in this thesis offers a scalable model for continuous improvement applicable to other industrial processes.