• Advertisements

Audits & Analytics – Know for Sure

Emily Sohn’s column Peanut Butter’s Checkered Food Safety History chronicles the recent history and scope of bacterial contamination and recalls focusing on four food products – peanut butter, cantaloupe, ground beef and spinach.  Each food stuff has its particular problems be it the rough rind on cantaloupe or the multiple animal sources of ground beef plus their own favorite pathogen problems with Salmonella, E. coli or Listeria.

She rightly points to the large scale extended and complicated food supply chain as a major contributor to the outbreak problems we face in today’s food market. Our experience suggests that best first line defense is the current Food Safety Management System model contained in standards and regulations such as ISO 22000, the Food Safety Modernization Act and the Global Food Safety Initiative.

These depend on a well defined system bolstered by analytics and real-time role specific reporting for timely and well informed decision making. This is summarized by a three step program:

  • Know your production and quality systems
  • Know your supply chain
  • Know for sure – verify with audits and use analytics to get maximum results from audits.

In our direct experience we have observed that integrating HACCP programs and SPC improves food safety results and in the case of ground beef purchase by the National School Lunch Program results in supply chains that delivered ground beef that never tested positive for the pathogens monitored. We also believe the additional step of extracting actionable information from audits with analytics helps maximize the food safety outcome.

Will these strategies eliminate all possibility of food safety events? No. Will they give us the best chance of a good outcome? Yes.


Surviving Close Encounters with the Committee

Food safety and quality management systems are a necessity if you are going to satisfy customer and regulatory requirements and meet corporate performance goals. While it is easy enough to review current best practices in food safety and quality systems and choose the options that will meet your compliance and performance needs, getting project sign off and budget approval from management committees can be an entirely different matter.

Much of that challenge comes from communication problems between two cultures, the technical and operational staff who develop the proposal and management who must understand it and sign off. The technical staff is most comfortable and fluent with the technical discussion. Management is fluent in the language of business and finance and evaluates the project in the context of corporate goals and ROI. Never were two groups more likely to have a failure to communicate.  

How do you reconcile the two groups to make the project happen?

Several proposal strategies have been used to bridge the gap and move forward to get the job done. The article “Justifying the Manufacturing Intelligence Project examines the most common strategies, where they work and how to maximize your chance of success.

Food Safety: Romney and Obama Focus on Different Solutions

In Christine Groman’s recent Scientific American blog, she reported on the presidential candidates responses to the food safety question in a larger questionnaire. The question was:

Question #7. Food. Thanks to science and technology, the United States has the world’s most productive and diverse agricultural sector, yet many Americans are increasingly concerned about the health and safety of our food.  The use of hormones, antibiotics and pesticides, as well as animal diseases and even terrorism pose risks.  What steps would you take to ensure the health, safety and productivity of America’s food supply?

The Republican response not surprisingly tended to minimize government and regulatory role. One of the secrets to the success we have enjoyed in the food safety arena is that there has been a long term collaboration between public and private sectors to produce the substantial increase in food safety levels we have observed. FDA and GFSI both deserve kudos. We should strengthen the abilities of all not minimize any.

The Democratic response focused on the changes in bacterial infections as shown in the chart below.

foodbourne bacterial infections 2010

It would be great if the food safety community can leverage this questionnaire to bring greater awareness the status of food safety programs. These data should be emphasized in public discourse so the public can see where we have had success and where we need improvement.

Perhaps we can even get to the point of explaining that active process management programs in conjunction with HACCP based food safety programs are at the heart of the modern FSMS as defined in ISO 22000, the GFSI audit schema and the Food Safety Modernization Act. John Surak and I first demonstrated that linkage in a 1998 article Integrating HACCP and SPCwhich examined the role of an SPC based process management program in maintaining safe food production in a Butterball Turkey plant. As we said about those findings, food safety is ultimate justification of high capability manufacturing. 

How Paper Based Traceability for the Food Industry Became “Non”-Legal and a Major Liability

Guest Blog by

Dave Miller, President
Mobia Soultions

When the Food Safety Modernization Act (FSMA) was first passed many initially interpreted it to mandate electronic traceability.  Given the industry push, the FDA’s often stated complaint about the ineffectiveness of paper traceability and the obvious preferences in the bill’s wording – that conclusion was understandable.  A closer reading showed that FSMA implied electronic traceability but stopped short of specifically requiring it.  But while FSMA did not directly specify the type of traceability technology required , it does require that the chosen technology work and work reliably – and that is where this story takes a sharp turn.

The concept of “efficacy” (working reliably) is an intrinsic part of every law and regulation that requires a specific function be performed. So, for instance, if you are required to keep grass below 6 inches high you don’t have to use a mower to keep the grass low, but if you are subsequently caught with high grass, then saying that you were using  “alien crop circles  to do it for you” would not be a valid defense. In the case of FSMA, there is no electronic traceability requirement per se, but your technology does have to work well and reliably – And now there are governmental and commercial studies are coming to light that prove paper traceability is not even close to meeting the efficacy standard.

In the absence of proof to the contrary – we as an industry have been assuming that manually searching of paper records for “words of interest”, such as lot numbers, item numbers, vendors, customers, etc.  , was slower, more costly, and a “little less accurate” than electronic searches. Until now that accuracy assumption had never been but to the test and we really had no idea of how much data was being missed in our manual searches.  Those tests have now been done – and the results ain’t pretty. There is now reliable proof that individual manual searches are only about 50% effective. 

To give you some idea of the problem, let’s do a simple test. Read the follow sentence one time, at your normal reading rate, and count the number of “f”s as you read.


Unless you went letter by letter counting, you likely found three, and even if you’ve found all six, most people don’t.  Now consider how well you would do if you were reviewing boxes and boxes of documents.

According to several governmental and private sector studies1,2, manual searches of paper documents were only able to find  40% to 60%  of “words of interest” known to be in the documents. And unlike these studies, in real world conditions we never know how many times the “words of interest”  appear in our documents so we never know how many were missed.

So what affect does a 50% find rate have in a manufacturing environment? If, for example a vendor told us they had shipped bad lots, but after searching all of our boxes of documents we correctly identify only 50% of the actually received vendor’s lots, then have the same “failure to find” rate for the manufactured lots made from those vendor’s lots  and then the ship to customers as well   – we are down to identifying a mere 12.5%  (50% * 50% *50%) of actual impacted shipments!

cascade of error rates








Even the 50% search loss only includes the losses encountered when searching correct, clean typed text pages.  Now assume the search was of hand written documents and included transcription errors. With imperfect data collection and documents the 12.5% effectively rate can easily be assumed to drop below 10%.  If other members of the supply chain used paper traceability as well, their loss of accuracy would be multiplied by this loss of accuracy.  So potentially a find rate as bad as  12.5% * 12.5% = 1.56%.

The FDA’s experience with paper seems to backup this grim finding. Until now paper traceability has survived because electronic solutions were outside of the practical cost range of many organizations. But the recent appearance of low cost, online inventory management systems, such as Mobia, effectively removes what was a valid claim by companies that they could not afford nor technically support an electronic system.

It is no secret that many companies are more motivated by potential business risks than criminal concerns. Here too, paper traceability is a significant liability.  Given the demonstrably poor performance of paper traceability vs. electronic traceability –  paper based systems cannot be reasonably called  “current good practices”. In the event of an “incident”, the damage caused by a food company’s inability to reliably track product due, to paper’s limitations, would dramatically increase the likelihood and size of “punitive damages”.   “Foreseeable Risk” may become you least favorite term.

Two other non-criminal concerns are insurance and commercial accreditations such as Global Food Safety Initiative (GFSI). Up until now insurance companies and underwriters have assumed traceability was their insured company’s internal matter. After recent contamination events, the insurance companies began to be concerned that their exposure may not be limited to just their insured’s activities but their insured’s entire supply chain. As such underwriters are now beginning to seek waivers or to make traceability demands on their insured’s entire supply chain.

As for GFSI and other accreditation organizations, they have always required that companies meet all pertinent food laws as a requirement to achieve their certification.  In other words, for US food companies, if you aren’t meeting FSMA requirements, such as effective traceability, your GFSI and other accreditations are at risk.

In the end, while FSMA didn’t directly mandate electronic traceability – it seems obvious that paper traceability is a functional failure.  At the very least companies with paper traceability will have an enormous disadvantage in proving compliance and efficacy.  And legal or not would you like to risk your business and freedom, let alone the safety of your customers and perhaps family, on a safety system that was 10% reliable?

(You can view a recording of Dave’s webinar, “The Impact of New Regulations and Standards on Packaging Traceability” )

1 Richmond Journal of Law and Technology Vol. XVII, Issue 3 ; New York Times

2 John Markoff “Smarter Than You Think” March 4,2011