Last year alone in America, the total value of quality defects was estimated to be $300 billion. This translates into literally millions of consumers who are unsatisfied with a product or service they paid for. In today’s competitive environment, businesses must adopt a proactive approach towards process improvement that can minimize the probability of defects through effective and efficient use of resources.

This goes beyond recognizing defects when they occur, and correcting them after-the-fact. This includes understanding the potential for defects so thoroughly – having such an intricate knowledge of what it is about your process that can produce these defects – that you are able to reduce the probability to virtually zero.

Overview: What is probability of defect?

Probability of defect is perhaps the most important indicator of a manufacturing process’s quality. It’s a statistical technique used to indicate how often a product will fail during its lifespan, which in turn allows you to estimate the likelihood that an item will result in customer dissatisfaction. It is calculated by analyzing all the failure modes that affect the product, determining which ones have the biggest impact on reliability, and then using statistics to measure how frequently those failure modes occur.

There are many different types of defects that can occur in Lean Six Sigma applications, using the probability of defect tool works particularly well on transactional processes with highly predictable outcomes — whether in manufacturing or service.

The main aim of investigating the probability of defect is to reduce the number of defects – which may be due to components, workmanship, or complete failure – and eliminate customer dissatisfaction; in other words, use it as a benchmark to determine whether or not products are performing reliably and consistently. If you find that you’re often unable to determine why something isn’t working in your manufacturing process, it’s time to analyze and measure your probability of defect.

3 Drawbacks to Probability of Defect

You should be aware of the following drawbacks to probability of defect.

1. Inaccurate forecasting

When you don’t take into account the number of products being inspected or the number being created, your estimate becomes inaccurate. If product creation increases, then that’s a lot more possible defects. If this is an issue for you, consider using other methods in addition to probability of defect, like failure mode and effects analysis (FMEA).

2. Questionable accuracy

If workers aren’t consistent in how they label defects, then all that data your defect inspection tool collects might not be very useful at all—inaccurate labels could mean an inaccurate calculation and incorrect actions taken by management trying to act on those results. And even if the labeling is consistent (which isn’t always easy across multiple shifts and languages), there may be more errors lurking under the surface than what is flagged during inspections—and thus more problems that go unaddressed because no one realizes they exist.

3. Not ideal for large amounts of data

When you’re dealing with very large amounts of information like thousands or millions of data points in your spreadsheet, calculating the probability of defect can be difficult or impossible because it relies on manual calculations and human judgment. If this is an issue for you and if not doing so would severely damage your business (which one assumes it would), consider outsourcing these calculations to someone who specializes in Six Sigma statistical analysis.

Why is probability of defect important to understand?

In any manufacturing process, the probability of defect is an important metric to track. This is because the probability of defect can have a direct impact on the quality of the final product. If the probability of defect is high, then it is likely that there will be more defects in the final product. This can lead to lower quality and may even cause safety issues. Therefore, it is important to understand the probability of defect in order to ensure that the quality of the final product is as high as possible.

An industry example of probability of defect

Despite how undeniably instrumental a positive customer experience is to driving revenues and reducing costs, businesses still fall short of delivering that. In fact, these failures are such regularly-occurring parts of business that they’ve become notorious-in-the-industry statistics.

These statistics are factual, documented, and the very straightforward result of what happens when organizations don’t take advantage of a resource like the probability of defect tool.

  1. Poor service causes 4 times more customer attrition than problems with pricing or products.
  2. You are exponentially more likely to have business from a repeat customer (60-70% probability) than get business from a new customer (5-20%).
  3. For every customer who complains, there are somewhere around another 26 who felt similar but remained silent.
  4. 96% of unhappy customers do not complain; 91% of them simply leave your business and never return.
  5. The average dissatisfied customer will tell between 9 and 15 people, on average, about their experience; 13% will tell 20 or more people.
  6. 70% of the overall buying experience is based on how the customer feels they are being treated.
  7. It costs 6 to 7 times more money to acquire a brand new customer than it does to retain an existing one.
  8. About 73% of the general shopping public remain loyal to brands because the customer service agents are friendly; conversely, 21% have reported leaving their favorite brands because they didn’t feel valued as a customer.

It is all about the experience, and that experience can be whatever it is you’re willing to give them; it should help knowing that, even in the case of defective products, all is not yet lost. Often it is the ability to respond positively in a challenging situation that the customer values the most. The defect still must be corrected, but there’s a good probability that world-class customer care will ensure you get the chance to do so.

4 best practices when thinking about probability of defect

There are four best practices to keep in mind when thinking about the probability of defects.

1. First, determine all possible sources of defects.

This includes manufacturing defects, design defects, material defects, and process defects.

2. Analyze and understand the impact of each type of defect.

For example, a manufacturing defect may cause a product to fail to meet specifications, while a design defect may cause a product to be unsafe.

3. Measure the likelihood of each type of defect occurring.

For example, manufacturing defects are more likely to occur in products with many components, while design defects are more likely to occur in products with complex designs.

4. Understand the consequences of each type of defect.

For example, a manufacturing defect may result in a product being returned, while a design defect may result in a product being recalled.

Frequently asked questions (FAQs) about probability of defect

What is Lean Six Sigma’s “accepted” defect rate?

As Six Sigma is a methodology for achieving near-perfect quality in any process or product, the goal is to have no more than 3.4 defects per million opportunities (or “DPMO”). This means that 99.9996% of all products or services produced by a LSS process are defect-free.

While the goal is always to reduce defects as much as possible, it is not always realistic or achievable to achieve a DPMO of zero. In many cases, LSS processes are able to achieve a DPMO of 1,000 or less. This is still an extremely high level of quality and is often more than adequate for most businesses.

What is the formula of probability?

In general, the formula for determining probability is:

Probability = Number of favorable outcomes / Total number of possible outcomes

For example, if there are 10 possible outcomes and 5 of them are favorable, then the probability of something occurring is 5/10, or 50%.

Are defect and defective the same?

No. A defect can be thought of as a potential problem that could cause a product to be defective, but it doesn’t necessarily mean that the product is defective. For example, if a product is missing a required component, that would be considered a defect. However, the product would only be considered defective if that missing component resulted in the product not meeting its intended purpose.

In other words, a defect is a potential issue that could cause a product to be defective, but it doesn’t necessarily mean that the product is defective. A defective product, on the other hand, is one that has anywhere from one to a million defects that have resulted in the product not meeting its intended purpose.

A “one size fits all” solution

Businesses are being pushed to deliver an ever higher level of quality, and to do so faster. The LSS probability of defect tool is a great way to make sure you’re staying on track to meet those goals because it helps you identify potential issues early on and address them before they become problems. It is one of the easier-to-implement LSS methodologies and is truly an effective tool for any business, big or small.

About the Author