Project details

Food safety and Risk analysis

Published by: Thusia

Status: CLOSED

Category: DATA INTELLIGENCE

Application domain: Food & Agriculture

Budget (EUR): FROM 15000 TO 30000

Project description

Companies dealing with testing and assessment of food quality before the small-mid-large scale retail trade have at their disposal a wide range of information on food quality and status, coming from experimental and laboratory tests, and results of certification processes. On this background, these are usually related to estimate contamination risks on the food to be retailed, by evaluating microbiologics, physics and chemicals parameters (among which allergens and pesticides, for example), and shelf life analyses (including pH monitoring, packaging and stocking conditions, salt and additives content in food, etc).

All these information gathered from testing are usually stored and processed by companies partially, or even neglected, due to the large amount of data to be considered, but actually represent an essential database to be exploited to extract several essential information as regards food safety, and hence the risks connected to the food processing from manufacturer to the retail trade.

At this purpose, Thusia want to introduce an innovative approach in processing all these data, based on a statistical algorithms and analysis capable of processing and classifying all these data to extract automatically correlations and trends, to acquire a higher awareness and confidence on food safety, and on the possible risks coming from food processing. In particular, two main tasks must be faced to define the statistical processing algorithm:

1) a classification and weighting of each variables ruling the phenomena of interest, for a correct and integrated risk analysis (that is anyhow prescribed as per the ruling regulations)
2) a set of criteria to classify as critical the available parameters and to correlate them with possible non-conformities occurring during the process (coming from suppliers, machinery malfunctions or blackouts, modifications or perturbations in procedures, allergenes or foreign body traced, etc).

The tool must be designed to process all these kind of data, and on the basis of the quality and safety quantitative criteria, to identify the critical parameters in terms of risk of food quality and safety degradation. Then, the algorithm must consider non-conformities and correlate them to out of range parameters, to identify critical operations or steps in the food processing. All these outcomes can be then provided to large scale retail trade companies, to verify, monitor, or defining new rules to their suppliers to guarantee food safety and prevent contamination and degradation risks.

At this purpose, all these source data must ideally be gathered in an unique environment (e.g. a database), to process them globally and get indicators and strategic information to assess risks on food safety.

Then, as a consequence, these can be partially or completely provided to the end users for forecasts and prevention actions. Hence, the output of the analysis tool must be flexible in order to be customized to end users and to suite their specific needs.

Project goal

The goal of the project is to provide a consultancy service to food production and retail companies aiming at maintaining and improving the hygiene/health FOOD SAFETY and QUALITY standards, through the information already gathered during the existing procedures and managing systems. The final goal consists in allowing the customers to MANAGE, CONTROL and MAKE FORECAST ON RISKS in a suitable way to finally increase the food safety of their products.

This goal must be pursued by developing a software tool based on STATISTICAL algorithms to process the data coming from experimental and laboratory tests on food and from inspection and traceability records, in order to find correlations and make previsions on their safety and RISK FORECAST. The algorithm must consider cross correlations between variables, and provide intuitive visualization of the results.