August 8, 2017 -
On August 14, the U.S. Food and Drug Administration issued guidance to clarify that a waiver to the Food Safety Modernization Act (FSMA) Sanitary Transportation of Human and Animal Food final rule (Sanitary Transportation rule) covers retail food establishments that sell food for humans, including those that sell both human and animal food, but does not apply to establishments that only sell food for animals.
Submitted by: Bill Bremer
August 8, 2017 -
In recent years, companies have been generating vast and ever-increasing amounts of data associated with business operations. This trend has led to renewed interest in predictive analytics, a field which focuses on analyzing large data sets to identify patterns and predict outcomes to help guide decision-making. While many leading companies use predictive analytics to identify marketing and sales opportunities, similar data analysis strategies are less common in occupational and process safety. Although the use of predictive analytics is less common in the field of safety, the potential benefits of analyzing safety data are considerable.
Just as companies are currently using customer data to predict customer behavior, safety and incident data can be used to predict when and where incidents are likely to occur. Appropriate data analysis strategies can also identify the key factors that contribute to incident risk, thereby allowing companies to proactively address those factors to avoid future incidents.
Predictive Analytics: In Theory
Let’s take a step back and look at what predictive analytics is and what it does. Predictive analytics is a broad field encompassing aspects of various disciplines, including machine learning, artificial intelligence, statistics, and data mining. Predictive analytics uncovers patterns and trends in large data sets for the purpose of predicting outcomes before they occur. One branch of predictive analytics, classification algorithms, could be particularly beneficial to industry, especially when it comes to avoiding incidents.
Classification algorithms can be categorized as supervised machine learning. With supervised learning, the user has a set of data that includes predictive variable measurements that can be tied to known outcomes. The algorithms identify the relationships between various factors and those outcomes to create predictive rules (i.e., a model). Once created, the model can be given a dataset with predictive variable measurements and unknown outcomes, and will then predict the outcome based on the model rules.
Predictive Analytics: In Practice
Like many in the transportation industry, this railroad had experienced a number of derailments caused by broken rails. Broken rail derailments can have particularly severe consequences, since they typically occur on mainline tracks, at full speed, and with no warning of the impending broken rail. Kestrel was asked to create a predictive model of track-caused derailments on a mile-by-mile basis to identify areas of high broken rail risk so the railroad could target those areas for maintenance, increased inspections, and capital improvement projects.
Penalized Likelihood Logistic Regression
As described above, classification models learn predictive rules in an original data set that includes known outcomes, then apply the learned rules to a new data set to predict outcomes and probabilities. In this case study, Kestrel used a logistic regression modified by Firth’s penalized likelihood method to:
- Fit the model
- Identify eleven significant predictive variables (based largely on past incidents)
- Calculate broken rail probabilities for each mile of mainline track based on track characteristics
The final model calculates a predicted probability of a broken rail occurring on each mile of track over a two-year period. The results suggest that the final model effectively predicted broken rail risk, with 33% of broken rails occurring on the riskiest 5% of track miles and 70% occurring on the riskiest 20%. Further, the model shows that the greatest risk reduction for the investment may be obtained by focusing on the 2.5% of track miles with the highest probability of a broken rail. This ability to predict where broken rails are likely to occur will allow the company to more effectively manage broken rail derailment risk through targeted track inspections, maintenance, and capital improvement programs.
Implications for Other Industries
The same general approach described in the above case study can also be applied to other industries—using KPIs to determine predictive variables and incidents as the outcome. The process is as follows:
- Measurements for defined variables would be taken regularly at each facility or unit. Precision increases as the measurements become more frequent and the observed area (facility/unit) becomes smaller.
- Once a sufficient number of measurements has been taken, they would then be combined with incident data to provide both the predictive variable measurements and the outcome data needed for training a model. This data set would be fed into a logistic regression or other classification algorithm to create a model.
- Once the model has been created, it can be applied to new measurements to predict the probability of an incident occurring at that location during the applicable timeframe.
Once predicted incident probabilities have been found, management would be able to focus improvement resources on those locations that have the highest probabilities of experiencing an incident. The classification algorithms also identify which factors have predictive validity, so management will know how improving those factors will affect the predicted probability of incidents occurring. In other words, they will know which factors have the strongest relationship with incidents, and can focus on improving those first.
Industrial companies are generating and recording unprecedented amounts of data associated with operations. Those that strive to be best-in-class need to use that data intelligently to guide future business decision-making.
The versatility of predictive analytics, including the method described in this case study, can be applied to help companies analyze a wide variety of problems. In this way, companies can:
- Explore and investigate past performance
- Gain the insights needed to turn vast amounts of data into relevant and actionable information
- Create statistically valid models to facilitate data-driven decisions
Submitted by: Will Brokaw
August 8, 2017 -
Designing and implementing a compliant Food Safety Management System (FSMS) can help organizations improve in many areas beyond the system’s defined tasks. It is critical for management to align the food safety objectives with the business needs for a successful and meaningful program implementation. Here are some of the top reasons why companies that work in the food industry may want to pursue developing and implementing an FSMS:
10. Identify and categorize the organization’s food safety risks.
Once this information is known, management can prioritize and decide how to eliminate or reduce business risks and liabilities to acceptable levels. These risks are often better controlled through strict management accounting. As a bonus, employees will become more attuned to thinking about risks and helping management improve overall operations.
9. Develop work instructions and/or procedures to guide employees’ actions and to ensure that each food safety task is completed in a disciplined manner and approved by management.
This will reduce the risk to an organization of an employee accidentally making a food safety mistake that causes the employee or others to be harmed (or worse). It also reduces the company’s risk of government inspections, fines, poor public perception, and loss of business due to a possible recall.
8. Assure management that they, in fact, know and understand the regulatory food safety requirements that must be met daily.
These requirements can be a driver of continual improvement by ensuring that the company has up-to-date procedures and work instructions for employees to follow every day.
7. Develop meaningful goals and objectives that drive food safety performance improvements and possibly reduce additional costs.
Each business will have different goals and these goals will likely change each year. Goals assure continuous improvement in food safety performance for the business over time.
6. Create a strong training and educational program that stems from well-written procedures and work instructions and that clearly defines the company’s requirements.
A well-trained workforce is a motivated and happy workforce. Turnover is reduced, accidents and incidents decrease, and production efficiencies increase. Employees are very aware when an organization takes time to ensure that each job requested is completed in the safest manner possible.
5. Develop appropriate monitoring and measurement practices.
Once all food safety requirements (e.g., FSMA, USDA, GFSI) are known and understood, the organization will be able to gauge food safety performance based on scientific data and regulations, and then guide the organization’s actions in a direction of continuous improvement and compliance.
4. Verify the FSMS is functioning as designed and implemented.
By continuously auditing each food safety program and function, the organization will discover issues of concern and non-conformances prior to an incident or agency/certifying body finding. Routine, non-biased audits allow the company to choose a timeframe that will help improve the situation without undue influence by outsiders.
3. Monitor and trend issues of concern and/or non-conformance and the actions used to rectify them through a fully functioning corrective/preventive action program.
As employees watch management fix problems, they will learn that management is concerned about continuous improvement. This will prompt employees to start making their own improvement suggestions. These suggestions will further drive improvement in areas outside the original FSMS.
2. Evaluate the business model and the FSMS in a holistic fashion.
By using this self-reflection and identifying improvement opportunities, management can direct responsibilities for improvement actions across many departments of the company. Each of these improvement opportunities has the potential to help the bottom line and reduce the possibility of a food safety liability now or in the future.
1. Know that the company has done everything to maintain the business in a manner that meets all food safety rules and regulations.
The last and most important benefit for an organization that goes through the process of designing and implementing a compliant FSMS is knowing that the organization has done everything possible to maintain its business in a manner that meets all food safety laws, regulations, and statutes every day the doors are open for business. To a business owner, that knowledge is priceless. This is how brands are built and how they maintain the promise of food safety to consumers.
Submitted by: Roberto Bellavia
August 8, 2017 -
Kestrel is pleased to be growing our resources to the food industry with the addition of Senior Consultant Melody Ge.
Melody brings a diverse background to the Kestrel team. She started her career in product development, including production and quality control of a vegan “chicken meat” product. She then transitioned to a Compliance Specialist at SQF Institute, where she established and developed the SQFI Compliance Program and maintained the integrity of the SQF certification; and developing the SQF Code.
Immediately prior to joining Kestrel, Melody served in a number of quality management and business development roles at Lidl, an international grocery chain. As the Deputy Quality Assurance Director, she oversaw suppliers, food safety control, and product quality monitoring and management to maintain quality and safety of product routine tasks.
At Kestrel, Melody will be serving as project manager for food safety-related projects. She will be supporting clients in developing and implementing GFSI schemes and supplier approval programs, and sharing her expertise in GFSI, FSMA, FSVP, HACCP, GMP, SQF, IFS, FSSC 22000, and ISO.
Melody holds a Master’s Degree in Food Science from the University of Maryland, College Park, and a Bachelor’s Degree in Food Science and Technology from Shanghai Ocean University, and is fluent in English, Mandarin/Cantonese Chinese, French, and German. She is a member of the Institute of Food Technology (IFT) and holds certificates in HACCP, Extrusion Processing and Technology and Commercialization, and Commercially Sterile Packaged Foods.
Submitted by: Bill Bremer
August 8, 2017 -
Companies grasp the importance of using technology to create business efficiencies. Integrating technology into traditional processes allows companies to stretch and empower limited resources. It offers ways to provide more value to company operations and management systems.
When it comes to technology integration, however, companies traditionally look for an isolated solution to a single problem—a find-it, fix-it approach. A simple example of this would be creating an Excel spreadsheet to manage data from multiple sources. While this creates an improvement beyond the traditional hard copy binder, it is a linear, isolated solution to one issue that offers minimal additional business value.
Consider the data on that spreadsheet and consider how business systems work. Does the data stand alone or does it impact other parts of the business? Does the business system operate in a silo or are there common elements with other business systems? In most cases, there is overlap between data, information, systems, platforms, etc. As a result, building a patchwork of technology solutions to address individual problems is only a short-term fix.
Truly valuable technology solutions take a relational approach that considers the immediate issue within the context of the overall business need, and then integrates multiple platforms/systems, as required, into an aligned system.
A forward-thinking, relational technology approach takes a solution perspective that thinks beyond the singular project need to the big picture and then designs backwards. It’s a shift in mindset from “How can I use technology to make this efficient?” to one that asks, “Ultimately, what does the big-picture, desired state look like…and how can technology get us there?”
A relational approach such as this follows these steps:
The following case study provides a real-world example of how a global chemical distributor is following these steps to create a relational technology solution that will improve business efficiencies across the company. Initially, this distributor wanted to pull data from facility reports for 150+ locations into one database—that was the “simple” problem. The old system had facilities entering data into Excel forms. That information was then pulled into Access so the data could be manipulated.
Understanding that the facility data is intertwined with many aspects of the business, Kestrel looked beyond this singular issue at the bigger picture. The forward-thinking solution would be to create a technology platform that would solve this facility data problem and could easily be expanded to other business needs, particularly since facility data is tied to most aspects of the business.
To do this, Kestrel built the facility form into SharePoint as the base application for the company’s overall system. SharePoint houses all data previously input into Excel documents for each facility broken up by 11 regional operating companies with multiple locations under each. The form requires that each facility contact fill out quarterly information on the facility (e.g., permits, fleets, transportation, personnel). Beyond the facility form, the SharePoint system currently has the following modules, which all feed into the facility form:
- Facility images
- Storage tanks
- Facility audits
The SharePoint system is continuing to be expanded to integrate other systems into a single source that will create significant business efficiencies. This approach is creating many benefits across the company:
- Data collection is easier and more accurate. There are no longer multiple, conflicting sources of facility information or requirements for multiple entry.
- The company is able to collect multiple levels of data and then associate that data to the individual facility or provide a composite report (i.e., data required for storage tanks, sustainability efforts, audits conducted).
- The look and feel of the forms in SharePoint are very similar to the original Excel documents, so it is an easy transition and very intuitive system to use. Little training has been required.
- The company can easily track information on all facilities. Management can export data to Excel and create reports. The company has complete ownership of data and deliverables.
- The system can create alerts for overdue items and generate real-time metrics and dashboards. Many additional options can be further customized based on ongoing business needs.
- Additional data from other systems being used across the company (e.g., auditing program) can be integrated and aligned into SharePoint as users become more familiar with the platform.
SharePoint is a dynamic solution tool that can be customized and designed to capture data and provide consolidated reporting to all levels of management. Because of SharePoint’s flexibility, the possibilities of what it can do are virtually endless:
- Creates a single, familiar platform that simplifies access
- Provides functionality for continual adaptation to meet future data management and reporting needs
- Adapts to the needs of the business, rather than the business adapting to the capabilities of the program
- Maximizes efficiency and connectivity between many field and corporate groups
- Allows information to be shared and tracked in multiple ways
- Allows users to easily create complex databases that are both manageable and flexible
- Gives the ability to manage sites/facilities/plants/departments for compliance purposes
- Simplifies the data entry process by providing user-friendly functionality
- Consolidates reporting
- Provides a dynamic solution – updates made to the tool are reflected immediately
- Allows local users to control and build sites to their specifications
- Allows all levels of users to work with it easily due to its intuitive nature
By having so many features and applications on a single platform, it is easy to tie them all together into an aligned system and to create multiple functions/uses for the data being collected from so many sources. With an aligned system, then, achieving the big-picture, desired state (rather than the short-term fix) becomes entirely possible.