Modeling and evaluation of quality monitoring based on wireless sensor and blockchain technology for live fish waterless transportation

• Published in: Computers and electronics in agriculture Vol. 193; p. 106642
• Main Authors: Feng, Huanhuan, Zhang, Mengjie, Gecevska, Valentina, Chen, Bingqi, Saeed, Rehan, Zhang, Xiaoshuan
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.02.2022; Elsevier BV
• Subjects: agriculture; Algorithms; bass; Blockchain; Correlation analysis; cost effectiveness; Cryptography; economic costs; Economic impact; electronics; Fish; Flexible wireless sensor; GA-BPNN-ARMA; model validation; Pollution monitoring; prediction; Quality assessment; Root-mean-square errors; Sea bass; Sensors; State estimation; Survival; Survival prediction; survival rate; Transportation; Wastewater; water pollution; Waterless transportation; Survival prediction; Flexible wireless sensor; GA-BPNN-ARMA; Waterless transportation
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2021.106642

•A flexible wireless sensor-based and blockchain monitoring system was designed.•GA-BPNN-ARMA algorithm can provide a reference for the live fish state diagnosis.•Critical parameters monitoring and prediction could reduce the fish quality risk. Live fish waterless transportation is a novel and cost-saving strategy, which can reduce the economic cost and wastewater pollution. However, it is still technology challenging to achieve intelligent monitoring and higher survival rate. In this paper, we presented a monitoring system for capturing quality-data changes in real time and verify captured information based on flexible wireless sensor and blockchain technology. The monitoring data was uploaded to the blockchain distribution platform for encryption, thus preventing the data from being tampered with. Quality evaluation applications were conducted for correlation analysis, state estimation and survival prediction by GA-BPNN-ARMA algorithm. Furthermore, the performance of the model, sensor, and blockchain were comprehensively analyzed and evaluated. In a field case scenario (live sea bass), dynamic monitoring and survival prediction were carried out for waterless transportation of >10 h, and the survival rate of sea bass reached about 89.5%. The mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) are 0.426%, 0.498% and 0.887%, respectively. This study provided an effective and reliable survival evaluation and transportation monitoring management.