Food safety undergoes a revolution: from systematized hygiene inspection to targeted inspection thanks to Data-Science

‍

Inspection companies can no longer operate under current conditions

Historically, companies in the commercial and institutional catering and food retail sectors have set up site inspection plans. 

This inspection plan includes checking that the site and its teams are capable of producing, serving and selling meals in line with the brand's quality standards, and without putting customers at risk of food poisoning.

This involves sending inspectors to each site, around 4 times a year. 

Each visit generally lasts between 30 minutes and 2 hours, during which the inspector goes through a checklist of hygiene points at all stages of work: 

• Storage
• Preparation
• Etc.

But then... 

This model requires considerable resources, which are increasingly difficult to find. 

This results in a significant ecological impact, a high cost for the customer compared to the value obtained, and paradoxically, rates that are too low to ensure the profitability of the service.

That's why the food industry and the companies that carry out these inspections are urgently calling this model into question. 

I refer you to the article "Why audits are no longer appropriate" for more details on this observation.

Faced with a lack of profitability and difficulties in recruiting and training inspectors, some ICT (Testing, Inspection and Certification) companies are forced to renegotiate their inspection rates upwards by at least 50%, for the same result as before. 

All this to get a quick overview, once a quarter, of what's going on at each site in its network.

The need for less volume and more quality

Catering companies are facing price inflation and labor shortages. 

However strategic food safety may be, their budgets are not compatible with a 50% increase in their hygiene inspection plan with no added value.

The solution?

Do less without increasing the risk of suffering a health crisis by poisoning a customer.

But how?

How can we reduce the number of inspections, which already represented less than 0.2% of average on-site presence over the year? 

Simply reducing the number of inspections would obviously increase the risk to your brand, which is already very high with such a weak management tool.

What makes this revolution possible: everyday digital data and data science

Real control of any activity depends on the intelligent use of sufficient data.

If I need to have confidence (or give confidence) in the critical, daily, standardized practices of my teams spread across a territory, it would never occur to any of us to train a quantity of inspectors to understand my standard, then ask them to get in their cars every day to give me a report on what they see, one day per quarter from each of my sites.

At a time when the exploitation of data has been demonstrated in virtually every professional sector, physically sending humans out - simply to tell me whether my daily practices seem to be respected - is a model from another era. 

Instead, it would be much more useful - and therefore fulfilling - for humans to go straight to the sites where doubt remains, or to think about how to rectify the deviances detected automatically, and to work towards achieving this.

Any CEO should ask his Operations, Quality and IT departments the following question:

• Our objective vis-à-vis our customers, our teams and our shareholders is to ensure on a daily basis that our practices are compliant and safe for our customers, with an efficient budget. On what sufficiently massive and reliable data can we base our risk management system?

• Then, how can we automate the use of this data to achieve this goal?

Until 2023, only the first question could be answered: digital self-check. 

But now, food safety too can be revolutionized by Data Science.

The fruit of many years of research and experimentation with our customers and partners, Eezytrace Score automates the processing of daily data. 

The restaurant and food retail sector now has, like all the other sectors that have already made their revolution, the data and tools to finally change this model of Quality surveillance through systematic physical inspection.

How Eezytrace Score enables effective monitoring with fewer inspections

Thousands of pieces of data are recorded and captured every day at each site. Firstly, thanks to the digitalized HACCP self-checks, but also to temperature sensors. 

Other relevant data can also be exploited, such as laboratory analysis results, Google Maps customer reviews, etc.

Eezytrace Score is the Data Science algorithm that :

‍

👉 Analyzes and detects the brand standard to be reached

👉 Identifies sites that deviate from the expected standard

👉 List the precise causes of risk for each deviant site

‍

The result is simple: the network's sites are automatically classified by risk level, saving countless hours of manual sorting and analysis. 

An analysis report is automatically generated for each site. Not only is this report based on daily data, it is also generated without any inspector having visited the site.

A site classified as not at-risk, which carries out its self-checks in a sound manner, and has compliant analysis results, has no reason to be inspected. This saves unnecessary budget.

Conversely, a site suddenly identified as risky needs to be given priority support. The budget for this support won't be spent checking whether it's at risk, and why. We already know that, thanks to data. 

The budget will therefore be used in a much more useful way: to anticipate.

And bring...

• Consulting
• Awareness
• Training
• Continuous improvement
• Handling non-compliances

In the case of ongoing (lower frequency) inspections, the inspector already has Eezytrace Score reports in his possession. He therefore already knows if there are areas that are not factually compliant (e.g. missing product traceability, undeclared recurring tasks). This means he can make better use of his visit time to check what he has doubts about.

‍