Risk Management Article

Ryan McKenna’s picture

By: Ryan McKenna

To date, this series focused on relatively simple data analyses, such as learning one summary statistic about our data at a time. In reality, we’re often interested in a slightly more sophisticated analysis, so we can learn multiple trends and takeaways at once and paint a richer picture of our data.

In this article, we will look at answering a collection of counting queries—which we call a workload—under differential privacy. This has been the subject of considerable research effort because it captures several interesting and important statistical tasks. By analyzing the specific workload queries carefully, we can design very effective mechanisms for this task that achieve low error.

Christopher Allan Smith’s picture

By: Christopher Allan Smith

This series is about planning for the worst that can face us.

It’s jumping-off point is the National Institute of Standards and Technology publication, “A Case Study of the Camp Fire—Fire Progression Timeline,” an epic and thorough study about the wildfire that changed the lives of my family, friends, and some fellow Quality Digest associates in November 2018. That fire razed most of the communities on the Paradise Ridge in Butte County, California, destroyed about 19,000 structures—95-percent of the residences in Paradise—and killed 85 people.

I have come to see my part in my community’s recovery as voicing the lessons we learned—literally taking the awful and searing things we learned that are of some use before, during, and after a disaster—and passing them on to other communities so they may face their trials with some better measure of success and safety.

Don Cox’s picture

By: Don Cox

Despite the high ratio of intelligent work-from-home (WFH) business professionals, the current cybersecurity landscape for that work model could best be described as disorganized and dysfunctional. Hackers have been busy exploiting these cyber risks, as evidenced from the reported 300-percent increase in cybercrimes in just the first quarter of 2020.

In the more than 791,790 cybercrimes reported throughout 2020, the total losses exceed $4.1 billion. For small or family-owned businesses, losses from a cyberattack could be unrecoverable and have ripple effects for years to come. The swift shift to remote work at the onset of the Covid-19 pandemic only exacerbated flawed and often stop-gap cybersecurity plans. Now, more than a year into virtual work for many Americans, it’s clear businesses can’t wait any longer to fully invest in cybersecurity for team members, programs, and education as WFH is here to stay.

Barbara Cuthill’s picture

By: Barbara Cuthill

The internet of things (IoT) offers many attractions for small and medium-sized manufacturers (SMMs) that may want to integrate IoT into their facilities and operations, or who seek to enter the IoT market with innovative products. However, when venturing into the IoT waters, it’s helpful to be prepared for the potential cybersecurity pitfalls, whether they are implications for organizational risk management when introducing IoT to the environment, or considerations for product design and support when entering the marketplace as a product vendor.

The NIST Cybersecurity for the Internet of Things program is working to provide the information that SMMs need to navigate these potentially turbulent waters.

Multiple Authors
By: David Darais, Joseph Near

In our last article, we discussed how to determine how many people drink pumpkin spice lattes in a given time period without learning their identifying information. But say, for example, you would like to know the total amount spent on pumpkin spice lattes this year, or the average price of a pumpkin spice latte since 2010. You’d like to detect these trends in data without being able to learn identifying information about specific customers to protect their privacy. To do this, you can use summation and average queries answered with differential privacy.

In this article, we will move beyond counting queries and dive into answering summation and average queries with differential privacy. Starting with the basics: In SQL, summation and average queries are specified using the SUM and AVG aggregation functions:

SELECT SUM(price) FROM PumpkinSpiceLatteSales WHERE year = 2020
SELECT AVG(price) FROM PumpkinSpiceLatteSales WHERE year > 2010

In Pandas, these queries can be expressed using the sum() and mean() functions, respectively. But how would we run these queries while also guaranteeing differential privacy?

Multiple Authors
By: David Darais, Joseph Near

How many people drink pumpkin spice lattes in October, and how would you calculate this without learning specifically who is drinking them, and who is not?

Although they seem simple or trivial, counting queries are used extremely often. Counting queries such as histograms can express many useful business metrics. How many transactions took place last week? How did this compare to the previous week? Which market has produced the most sales? In fact, one paper showed that more than half of queries written at Uber in 2016 were counting queries.

Counting queries are often the basis for more complicated analyses, too. For example, the U.S. Census releases data that are constructed essentially by issuing many counting queries over sensitive raw data collected from residents. Each of these queries belongs in the class of counting queries we will discuss below and computes the number of people living in the United States with a particular set of properties (e.g., living in a certain geographic area, having a particular income, belonging to a particular demographic).

Graham Freeman’s picture

By: Graham Freeman

Here’s an unfortunate truth: The story of the Covid-19 pandemic is one of epic quality failures in almost every area imaginable. Although there have been some admirable successes, such as the food and beverage organizations that have ensured the continued safe delivery of food supplies to most regions, failures both large and small have caused an untold amount of damage to the infrastructure of society and business. Arguably, these quality failures have worsened the impact of the pandemic, including economic devastation and even a higher death toll.

Here are just a few of the quality failures that will become prominent themes in the Covid-19 narrative.

Multiple Authors
By: Bob Holmes, Knowable Magazine

This story was originally published by Knowable Magazine.

Most of us won’t soon forget that disconcerting moment last spring when grocery store shelves were suddenly bare where the flour, pasta, and other staples should have been. The news told of farmers dumping milk—nearly four million gallons a day, by one account—smashing eggs, and euthanizing chickens that they couldn’t get to market. Crops worth billions of dollars were wasted, some rotting in the field, as restaurants and other food service businesses, shuttered by lockdowns, stopped buying.

The problem was short-lived, fortunately, as growers pivoted to new buyers, shippers and packers adapted, exports resumed, and the food system—the complex web of players that move food from farm to fork—came back to life. “Overall, the food system has been quite resilient,” says Johan Swinnen, director general of the International Food Policy Research Institute, a leading international think tank. “It’s hard to imagine a bigger shock than we’ve had now. And despite that, if you look at the rich countries, even countries like China, the food supply has not been a problem almost anywhere.”

Rita Men’s picture

By: Rita Men

Ending the pandemic depends on achieving herd immunity, estimated at 70 percent or even 80 percent to 90 percent of a population. With some 30 percent of Americans telling pollsters they have no interest in getting vaccinated, that’s cutting it a bit close. The numbers are even worse in many other countries.

Del Williams’s picture

By: Del Williams

On conveyor systems in the food processing industry, some powdered and bulk solid materials such as grains, sugar, and creamer are ignition-sensitive in specific concentrations, particularly when exposed to static electricity discharge. Key concerns are conveyor-system connection points such as inlets, outlets, and storage bins. The concentration of dust can become sufficiently high for a deflagration to occur with accidental exposure to an ignition source, such as static electricity, a spark, flame, or even high heat or friction.

So, the characteristics of the material conveyed and the type of conveyor along with its associated component parts and connection points should be considered in the system’s design to avoid a serious risk of dust combustion and explosion. By carefully selecting and integrating the conveyor system and its components, food processors can minimize the risk of dust explosions while safely conveying materials in a hygienic and energy-efficient manner.

Syndicate content