Authors: Chikhale Sahil Chandrakant, Professor Abhishek Popat Bangar, Professor Bhosale S. B., Dr. A. A. Khatri
Abstract: Mangoes are sold ripe, but ripeness is usually judged by colour and smell, and in many markets it is forced with industrial calcium carbide, which is illegal and leaves toxic residues. Sensing ripeness and detecting carbide with low-cost connected devices, termed here IoT-based smart mango assessment, is widely proposed as a remedy for small traders and inspectors. This paper presents a focused narrative review that synthesises five studies: a lab trial that fused an electronic nose with an acoustic firmness sensor on Harumanis mangoes, a low-cost MQ-sensor electronic nose with regression for mango sugar and hardness, an ESP32 IoT electronic nose with a dashboard tested on bananas, a visible and near-infrared spectral sensor for detecting carbide-ripened mangoes and bananas, and an electrochemical biosensor for calcium carbide. The review examines three connected themes: the physiology and chemistry that set what can be sensed, the strength of the empirical evidence, and the conditions that matter for deployment. In the first study, an electronic nose alone separated only four of six ripeness groups, while the fused system separated all six, and a neural classifier rose from 66.7% to 84.4% accuracy. The MQ-based nose predicted mango sugar content with a correlation of 0.815 and hardness with 0.847. The spectral sensor classified mango ripening method with 95% accuracy but caught only 0.67 of carbide-ripened mangoes. The biosensor reached a detection limit of 0.6 nM. No study combines ripeness sensing, carbide detection and IoT reporting on mangoes in one tested device. The review contributes a comparative table, a conceptual architecture for a connected mango ripeness and carbide screening unit, and a prioritised agenda of research gaps.
