Mira Trebar

RFID Data Loggers in Fish Supply Chain Traceability

Radio frequency identification (RFID) is an innovative and well-recognized technology that supports all kinds of traceability systems in many areas. It becomes very important in the food industry where the electronic systems are used to capture the data in the supply chain. Additionally, RFID data loggers with sensors are available to perform a cold chain optimization for perishable foods. This paper presents the temperature monitoring solution at the box level in the fish supply chain as part of the traceability system implemented with RFID technology. RFID data loggers are placed inside the box to measure the temperature of the product and on the box for measuring ambient temperature. The results show that the system is very helpful during the phases of storage and transportation of fish to provide the quality control. The sensor data is available immediately at the delivery to be checked on the mobile RFID reader and afterwards stored in the traceability systems database to be presented on a web to stakeholders and private consumers.

Rfid-based traceability along the food-production chain [Wireless Corner]

This contribution explains and analyzes the use of RFID (radio-frequency identification) for defining a complete traceability system applied to the food-production chain. The paper contains a summary of the actual work developed to test the ability of radio technologies to perform traceability at different food companies in a variety of sectors: wine, fish, and meat. Each pilot experience is explained, with special emphasis on the radio segment implemented by RFID technologies and sensors, whether connected by wired or as elements of a wireless sensor network. The application of the new RFID-based system at the three investigated sectors, and the return on investment that the companies could obtain by its usage, are the core of the paper.

Use of MATLAB Neural Networks Toolbox in a character recognition problem

We present the use of the MATLAB Neural Network Toolbox (NN Toolbox) in simulations of neural networks. We suggest ways for undergraduate students to solve a character recognition problem with feed‐forward neural networks. The software provides the user with a very simple way to define several neural network architectures with different parameters. The solution of the character recognition problem is described from the beginning: collecting the data, data encoding, defining the input‐output mapping architecture to the training, and testing the neural networks with the NN Toolbox.

Evaluation on home storage performance of table grape based on sensory quality and consumers’ satisfaction

With continuous rise of table grapes consumption and increased public awareness of food safety, the quality control of grapes in storage after purchase is not sufficiently examined. Home storage constitutes the last and important stage in grape supply chain. Literature review shows that few researches on grape quality focus on the home storage stage compared with numerous researches reported on the quality control during postharvest and transportation process. This paper reports the performance evaluation of grape quality at home storage and consumers’ satisfaction using integrated sensory evaluations. The internal attributes, including Texture, Taste and Odor of the table grapes and the appearance indices, Color and Cleanliness are examined. Key results show that during home storage, all the internal attributes decrease rapidly as time goes on, and cleanliness and color appear to be deteriorating in a lower speed. A comprehensive quality index was created to measure the quality of table grape which has high correlation with the Overall acceptability perceived by consumers.

Improvement of Traceability Processes in the Farmed Fish Supply Chain

NSFC – National Natural Science Foundation of China K. C. Wong Education Foundation (Hong Kong) Springer-Verlag

Adaptive Modularity and Time-Series Prediction

This paper focuses on the use of recorded time-series to estimate future values as a function of their past values. We study the local events in input space and apply them as classes of similar patterns to the problem of short-term prediction. The decomposition of the time-series into the patterns formed from d past values denoted as an input vector and the true future value in the observed time-series is performed. From the observation of past values we can conclude that similar input vectors often have similar predictive values. We assume that this principle can be expanded in the future. The time-series is based on a similarity measure partitioned into similar patterns grouped into classes. Each of these classes computes the predicted value. The final predicted value is then determined with only one class obtained by the classification from the present input vector.

Cold Chain and Shelf Life Prediction of Refrigerated Fish – From Farm to Table

Fresh perishables are normally stored and distributed with a proper cold chain control in the supply chain from farm to retail. Usually, the consumers break the cold chain after the point of sale. The question is whether consumers are aware of requirements during the transport to and storage at home. The handling conditions and temperature changes can significantly decrease the shelf life and cause faster spoilage of food. The study presents two examples of shelf life prediction. The first one is based on temperature measurements of fish covered with ice in a Styrofoam box with supported information of environment temperatures in the cold store, uncooled car and refrigerator. In the second, measurements from first phase of storage on temperatures (0 °C–4 °C) were used with assumption of fish stored later on higher temperatures without ice. The results show important shortening of shelf life after the point of sale.

Predicting Mechanical Properties of Rubber Compounds with Neural Networks and Support Vector Machines

The quality of rubber compounds is assessed by rheological and mechanical tests. Since mechanical tests are very time consuming, the main idea of this work is to quest for strong nonlinear relationships between rheological and mechanical tests in order to reduce the latter. The multilayered perceptron and support vector machine combined with data preprocessing were applied to model hardness and density of the vulcanizates from the rheological parameters of the raw compounds. The results outline the advantage of proper data preprocessing.

Reinforced Search in Stochastic Neural Network

A reinforced search algorithm for the stochastic feedforward neural networks is described. A stochastic neuron is used in a network as a searching unit. Reinforcement signal from environment is used for weights and variance adaptation. This is experimentally compared with more traditional techniques like gradient-based learning algorithm and evolutionary algorithm.