Copyright © 2026 Bonfiglioli Consulting. All rights reserved. P.I. 02646871208 Privacy Policy | Cookie Policy | Adjust Preferences
In this article, you'll find everything you need to understand and apply FMEA in your company. Let's start with the difference between reactive management and structured defect prevention, explaining why waiting for a problem to arise always costs more, both economically and in terms of reputation. We will then look at what FMEA concretely is (Failure Mode and Effects Analysis), how the IPP (Indice di Priorità del Rischio - Risk Priority Index) is calculated with a practical example, and who should be involved in the analysis to make it truly effective. You will also find three real-world case studies from different industrial sectors, guidance on when is the right time to start an FMEA, and how this tool integrates into the Lean system along with Poka-Yoke and continuous improvement.
How many of the defects you handle every week were predictable? Probably more than you think. Not because skills are lacking, but because a structured method for risk analysis is missing. First that turn into problems. This method is called FMEA.
In many companies, quality management works like this: a defect arises, it's analyzed, and measures are taken to fix it. A costly, exhausting, and often useless cycle, because the same problem tends to reappear, perhaps in a different form, at another stage of the process.
The Failure Mode and Effects AnalysisFailure Mode and Effects Analysis — Failure Modes and Effects Analysis) break this cycle. It's not a problem-solving tool: it's a tool for prevention of the problem. Its underlying logic is simple: instead of waiting for failure to happen, one reasons systematically about come it could happen, why, and with what impact—before it's too late to intervene economically.
The practical difference between the two cultures is measured by costs: addressing a problem during the design phase costs enormously less than correcting it during production, and infinitely less than managing it after it has reached the customer.
For each stage of a production process, or for each component of a product, the FMEA Ask three fundamental questions:
How could this phase go wrong? il Failure mode)
If it goes wrong, what consequences does the customer or the downstream process suffer? the effects)
Why might this happen, and would we notice it in time? the reasons and the existing controls)
“In this context, ”fault" doesn't just mean breakdown: it includes any anomaly, error, non-conformance, or malfunction that causes a planned function to fail. A worn gasket that no longer guarantees a seal. A bolt torqued insufficiently, causing vibrations. A process parameter out of control, intermittently generating scrap.
The value of FMEA is not in the list of possible failures – anyone can do that after years of experience. It lies in the method for weighing them and for deciding where to act first, in a rational and team-shared manner.
Once the possible failure modes have been identified, the team assigns each one three numerical values on a scale from 1 to 10:
Gravity (G): How severe is the impact on the customer in case of a failure?
Probability (P): How likely is the cause to manifest?
Detectability (R): How difficult is it to notice the problem before it reaches the customer?
The product of the three values — G × P × R — generate the’IPR (Risk Priority Number), also known as RPN in international literature. The higher it is, the more critical and prioritized the risk is.
A concrete example: an aesthetic defect on a panel (low severity), which occurs rarely (low probability) and is visible to the naked eye (high detectability) will have a very low RPN, meaning it can be managed without urgent actions. A failure of a safety component (high severity), caused by an unstable process variable (medium probability) and which is not discovered until final testing (low detectability): in this case, the RPN will be high and requires immediate action.
The team establishes an acceptability threshold (typically around 70 out of 1000) and works to reduce the IPR of all critical issues that exceed it, by addressing the cause, process, or control systems.
FMEA is not a document to be filled out alone. It is, by its nature, a interdisciplinary toolrequires design, quality, production, logistics, and maintenance to be around the same table.
Each function brings a vision that others do not have. The designer knows why a certain sizing was done. The production manager knows where the process is unstable. The maintenance technician knows about recurring failures despite repairs. The quality manager has access to the history of non-conformities and complaints.
Only by cross-referencing these perspectives does the analysis become solid. And there's a side effect that often surprises those doing FMEA for the first time: the team learns to see the process with fresh eyes, to speak the same risk language, to share a responsibility that was previously fragmented by function.
The methodology comes to life in concrete results. Here are three real-world applications that demonstrate the versatility and impact of FMEA in diverse industrial contexts.
During the development of a new critical component—a heating resistor integrated into the water pump—the team applied FMEA prior to production launch. By cross-referencing the functional analysis with historical yield and complaint data from similar products, they identified three sources of serious risk that no one had considered yet. The result was the redesign of two components and the redefinition of the brazing process specifications—all before the product reached the assembly lines.
In a company that manufactures tire repair machinery, FMEA was introduced as a systematic methodology during the development of a new product. The project combined training and practical application: the team learned the methodology by working directly on their own process, not on a hypothetical case. An approach of learning by doing which has accelerated both the adoption of the tool and the quality of the analysis produced.
In a development project in the work vehicle sector, FMEA was applied to 39 functional groups, according to the client's procedures. At the end of the project, the client noted concrete benefits: higher product quality, better control of process risks, a consolidated preventive approach, and significantly more efficient control plans compared to previous projects.
Three different sectors, three different product types, the same underlying logic: Anticipating the problem is always more valuable than correcting it..
The ideal answer is: the as soon as possible. The maximum value is obtained by applying FMEA before designing a new process or launching a new product, when there is still room to intervene in the design without prohibitive costs.
However, FMEA is also useful for processes already in production, especially in the presence of recurring defects that cannot be definitively eliminated, or when a significant change is introduced to the process or product. In these cases, the actions that emerge are corrective, but the method is the same: systematic, structured, root cause-oriented.
FMEA is not a one-time exercise. It is a living document, to be updated every time the process changes, every time an unexpected failure emerges, every time a corrective action is implemented. Organizations that see results from FMEA don’t just set it aside—they keep it up to date.
In Lean Thinking, defects are among the main sources of waste (change): every non-conformity that goes through the process without being intercepted generates costs without creating value for the customer. FMEA is the tool that Tackle this waste at its root, even before it manifests itself.
It works in synergy with other Lean tools: it helps to understand where a device Mistake-proofing is it really necessary, which phases require more robust statistical controls, where to focus projects on continuous improvement. It is not an alternative to other approaches: it is the starting point to make them more targeted and effective.
FMEA, like all Lean tools, reveals its full potential only when applied to real cases, with a team, under the guidance of someone who has used it for years in industrial contexts. For those who want to take this step, Lean Factory School® designed the course “FMEA is Easy!”a day with a high practical content, with the motto that guides the whole school, Learn by doing.
Theoretical concepts account for less than 20% of the time; the rest is devoted to exercises, discussion, and direct application. You leave the classroom with a tool that’s ready to use, not a textbook to study.
FMEA stands for Failure Mode and Effects Analysis, In Italian, “Failure Modes and Effects Analysis.” It is a structured risk analysis methodology used in industrial settings to systematically identify all possible ways a process or product could fail, evaluate its consequences, and intervene preventively before the problem occurs.
The Product FMEA (or DFMEA, Design FMEAis applied during the design phase and analyzes potential failures related to design choices: materials, tolerances, geometries, functions. The Process FMEA (o PFMEA, Process Failure Mode and Effects Analysis) instead applies to production phases and analyzes how errors or variability in the manufacturing process can generate non-conformities. In complex projects, the two analyses are developed sequentially and feed into each other.
The RPN (Risk Priority Number), also known asRisk Priority Number) in the international literature, it is calculated by multiplying three values, each on a scale of 1 to 10:
IPR = G × P × R
The theoretical maximum value is 1000. Typically, an acceptability threshold is set around 70: all failure modes with a higher Severity require priority corrective or preventive actions.
FMEA is mandatory in several regulated contexts. In the automotive sector, standards IATF 16949 and the manuals AIAG-VDA they explicitly require it as part of the APQP processAdvanced Product Quality Planning). In the aerospace sector, it is among the requirements AS9100. In the medical sector, it is required by regulations ISO 13485 and from the FDA's guidelines for medical device risk management. Outside of these regulatory contexts, FMEA remains strongly recommended as a best practice in quality and prevention.
It depends on the complexity of the process or product being analyzed and the team's experience. For a medium-complexity manufacturing process, a complete FMEA analysis typically requires between 8 a.m. and 4 p.m. In group work, distributed over multiple sessions. Very complex processes (as in the case described in the article with 39 functional groups) can take weeks. The time invested in FMEA is, however, always less than the cost of managing defects once production has started or, worse, after they have reached the customer.
Absolutely yes. FMEA is scalable: it can also be applied to a single critical process with a team of 3-4 people. It does not require expensive software (you can start with a structured Excel sheet) and the return on investment, in terms of defect reduction, is measurable even after the first applications. Indeed, in SMEs, FMEA can have an even more visible impact, because it is often the first structured risk analysis tool that the team uses.
The Failure Modes, Effects, and Criticality Analysis (Failure Mode, Effects, and Criticality Analysis) is an extension of FMEA that adds a quantitative analysis of failure criticality, based on reliability data and failure rates. It is primarily used in high-reliability sectors such as aerospace, defense, and nuclear. “Classic” FMEA is more agile and widespread in general manufacturing, as it does not require detailed statistical reliability data to be effectively applied.
Copyright © 2026 Bonfiglioli Consulting. All rights reserved. P.I. 02646871208 Privacy Policy | Cookie Policy | Adjust Preferences