Towards smart fruit dryers: Integrating machine vision and response surface methodology for multi-objective monitoring and optimization of kiwifruit drying process

This study aimed to monitor and optimize the kiwifruit drying process in a combined convection–infrared hot air dryer using computer vision and Response Surface Methodology (RSM). Drying experiments were performed by varying inlet air temperature (50–70 °C), air velocity (1–2 m/s), and infrared power (200–300 W). Key performance indicators included drying time, specific energy consumption, energy efficiency, and quality attributes such as firmness, color change, and shrinkage, with the latter two measured via computer vision. RSM optimized the process to 62.75 °C air temperature, 248.28 W infrared power, and 1 m/s air velocity. At these conditions, drying time, energy consumption, energy efficiency, color change, shrinkage, and firmness were 64.87 min, 27.70 kWh/kg water, 16.92 %, 5.55 %, 38.01 %, and 19.16 N, respectively. The results highlight the potential of computer vision for real-time quality assessment and suggest a pathway toward developing smart, non-destructive, and automated drying systems for fruit processing industries.