Conference Paper

Environmental variables traceability device to predict postharvest quality and remaining shelf life of fruit and vegetables

Book Series cp
Acta Horticulturae
  • Volumen: 1311
  • Fecha: 04 June 2021
  • Páginas: 209-215
  • ISSN: 24066168 05677572
  • Source Type: Book Series
  • DOI: 10.17660/ActaHortic.2021.1311.26
  • Document Type: Conference Paper
  • Publisher: International Society for Horticultural Science
© 2021 International Society for Horticultural Science. All rights reserved.Actual consumers demand nutritive horticultural products with good quality and safety. Therefore, the industry must guarantee these properties throughout the complete supply chain, where environmental factors affecting quality are not usually the recommended ones. Temperature, relative humidity, and several gases (O2, CO2 and C2H4) partial pressures, are the most important factors to preserve quality and safety. Therefore, knowing such information will lead to making decisions to achieve a better quality of the commodities and reduce food losses. This research aims to develop a useful system able to monitor variables and predict the shelf life during the postharvest storage. The innovation system presented consists of a wireless sensor network able to connect the measuring devices to several independent routers using Wi-Fi technology. Since, the measuring devices can send the information to different routers placed in different stages of the supply chains, access to the recorded data are guarantee offering high communication flexibility. The device is designed to avoid the main disadvantages of the different technologies available in the market, as ZigBee and RFID systems which need proprietary receptors to connect to the Internet and send the information recorded. The developed devices can record the information when WIFI signals are not available. They send the information as soon as an access point signal is in range. To solve the availability of WIFI signals in some stages (like transportation), a portable gateway node has been developed. A methodology for building shelf life models using data recorded in several stages of the supply chain has been proposed. A logistic regression model using machine-learning techniques is proposed in this paper in order to predict the shelf life.

Author keywords

    Indexed keywords

      Funding details