What Is Root Cause Analysis? Definition, Methods, and Real-World Examples

Root cause analysis is a data-driven, systematic approach to diagnosing equipment issues. The approach is widely used across industries. By identifying the source of failure, root cause analysis, or RCA, drives more efficient maintenance practices, increases productivity, and keeps overhead low. 

In this blog post, we will explore why RCA is so critical. We will talk about some of the most important RCA tools and discuss how to leverage data analytics to improve outcomes. 

Understanding Root Cause Analysis

Root cause analysis is an umbrella term for analysis approaches aimed at uncovering the source of failures. Instead of just treating the symptoms of the problem, RCA digs beneath the surface to learn what caused the problem in the first place. In some cases, this may involve root cause oscillation diagnosis, where recurring issues are traced back to cyclical patterns in equipment behavior or operating conditions. 

Root cause analysis is premised on a simple idea: Determining the root cause of problems will open the door to a long-term solution that will prevent the issue from recurring.  

Key Methods of Root Cause Analysis

The purpose of root cause analysis is to prevent failures from recurring. RCA aims to pinpoint the underlying causes of the issue, instead of simply treating the symptoms of the problem. 

Here are the most popular methods.

The Five Whys Technique
Asking why may not sound like a sophisticated analytical method, but when implemented correctly, the five whys is a powerful RCA tool. 

Begin by identifying the problem at hand and ask why this problem has occurred. Continue to query as you learn more about the problem.  

Imagine that your HVAC system is down. 

Why did the HVAC system fail? The system failed because it overheated. 

Why did the system overheat? The system overheated because a fan broke down. 

Why did the fan break down? The fan broke down because its wiring was faulty. 

The method is called the ‘Five Whys,’ but the number of questions isn’t fixed; you may need fewer or more than five to reach the root cause.

Fishbone (Ishikawa) Diagram
The fishbone diagram is a classic RCA tool, popular for its structured, visual breakdown of contributing factors. It is also known as the cause-and-effect diagram, or the Ishikawa diagram, after its inventor, Kaoru Ishikawa.  

As its name suggests, the fishbone diagram uses the outline of a fish, complete with bones, to visualize the root causes of machine failure. 

In a fishbone diagram, the main failure event is represented by the fish head and spine. The fish bones branching off the spine all represent different categories of problems that led to that main event. 

The categories included in a fishbone diagram typically include personnel issues, problems with tools or components, larger equipment issues, and problems with maintenance strategy. Smaller bones, branching off of those categories, can fill in more details and critical information. 

The value of a fishbone diagram is its ability to group issues into key categories. This helps subject matter experts (SMEs) to organize their thoughts and plan the next steps more easily. It is also a great way to see, at a glance, which facets of the process are most likely contributing to the problem, so the team can prioritize investigation and focus resources where they will have the greatest impact.

Pareto Analysis
Pareto analysis is based on the Pareto principle, which was developed by Italian economist Vilfredo Pareto in 1906. The principle maintains that most events are caused by a very small group of catalysts. In other words, 80% of machine failures are caused by 20% of triggering events. 

Pareto analysis uses Pareto charts to identify the most frequent causes of equipment failure. A Pareto chart combines a bar graph and a line chart to reveal which issues contribute most to overall failures. Once the most common sources are uncovered, you can allocate maintenance resources more effectively. 

Here is how a Pareto chart works: 

1. Identify and list the major causes of equipment or process failures over a defined time period (e.g., one year). 
2. Count how often each cause occurs during that period. 
3. Plot the causes on a bar chart in descending order of frequency, with the height of each bar showing how many times each cause occurred. 
4. Calculate the percentage of total failures attributable to each cause, and then add a line graph on the same chart to display the cumulative percentage. 

This process makes it clear which issues cause the most problems for your organization. By addressing these root causes, you can correct failures and reduce the chances of them happening again. 

Comparing RCA Methods

Which RCA method is best for you? It depends on the problem at hand and the resources your organization brings to bear. 

The five whys approach is often the best choice for straightforward problems with a single, well-defined root cause. If you are dealing with human error, such as incomplete preventive maintenance tasks, you can pinpoint exactly where the problem is located so that you can correct it. 

The fishbone diagram is often best when you are facing multiple issues. Perhaps you are dealing with both human error and raw material problems. The fishbone approach will show you where the issues are concentrated. 

The Pareto analysis gives you a big-picture view of the larger patterns in your organization: Where do issues tend to arise? Which trends should you be on the lookout for? Where should you focus your labor power and your resources? 

Ideally, organizations will use a combination of all of these methods to successfully increase reliability and productivity. 

Five Steps in the RCA Process

Defining the Problem
At this stage, clearly articulate what the problem is. For example, maybe you are experiencing a leaky pump or a fouled heat exchanger.  

Data Gathering Techniques
Collect data from multiple sources, such as process data historians, quality databases, maintenance work order systems, and alarm databases.  

Seeq makes it easier to access data for analysis. It connects directly to data sources like process data, quality data, and MES data, allowing you to use the information right away. Seeq AI Assistant helps teams test ideas and analyze data quickly, so you can gather information and uncover insights faster.  

Identifying Causal Factors
Drill down into your data to pinpoint the factors behind your problem. Remember that you may have multiple, seemingly unrelated factors that all contribute to the same problem. 

Pinpointing Root Causes              
Follow each of your factors back to the root cause. For example, if you have identified “overheating” as a factor behind motor failure, work backward to find out what is causing the overheating.  

You can use Seeq to narrow down root causes by focusing on specific time periods and comparing data with a wide range of analysis tools. Once you identify a root cause, you can set up alerts that send you email notifications if the issue occurs again. You can choose methods ranging from simple Capsule definitions to more advanced options like condition-based monitoring, trend forecasting, regression modeling, soft sensors, golden profile monitoring, and anomaly detection with Seeq’s point-and-click ML tools. 

Implementing Preventive Measures
Now that you have identified the root causes of your issue, put measures in place to prevent recurrence. Create clearly defined workflows and assign tasks to specific people to ensure accountability and reliability. 

Challenges and Limitations of RCA

Like any other strategy, RCA requires some training to get it right. Using the right tools can help teams avoid the pitfalls of this approach. 

Potential for Inaccurate Results
Inexperienced teams, or teams working without proper tools, can sometimes get inaccurate results. A lack of proper data can also drive incomplete or incorrect analysis.  

Ensuring Proper Execution
Training your teams in the proper methodology is crucial. Creating standardized workflows also makes a difference. Using a platform like Seeq to collect and analyze data helps you find real-time insights that you can act on immediately. 

Real-World Applications of RCA

Root cause analysis is widely used in just about every industry, from mining and engineering to health care and manufacturing. 

Engineering teams use RCA to understand the causes of unexpected downtime, such as compressor failures or products out of specification. For example, chemical companies may use RC analysis to monitor abnormal conditions around critical rotating machinery before failures occur, thereby avoiding potentially millions of dollars in repair costs. Using a platform like Seeq, each process signal is monitored for anomalies, enabling engineers to focus their investigation efforts and make rapid decisions. Over time, teams can analyze and identify early indications of failures and apply the same anomaly detection model quickly and effectively for other assets.  

There are countless examples of RCA being used to optimize productivity across sectors. In every case, RCA has the same purpose: to assess what went wrong in a particular process so that the organization can make the necessary changes and stop that problem from happening again.  

RCA in Manufacturing

Manufacturing and maintenance teams use RCA to uncover the causes of production failures, equipment breakdowns, and unexpected faults in critical machinery. By improving machine availability, optimizing performance, and boosting efficiency, RCA helps operations run more reliably. 

RCA is driven by precision and advanced analytics, often powered by IoT data and modern software. Time-tested techniques such as graphs and charts remain valuable, but platforms like Seeq add even more capability by simplifying analysis and documenting each step in a journal. This makes it easy to share findings and keep the RCA process consistent, fast, and effective. 

Start Moving Beyond Reactive Problem-Solving

Effective data analytics can transform RCA by speeding up the process and improving the accuracy of results. Seeq makes this possible by helping you uncover patterns and issues across your organization. And the Seeq Industrial Enterprise Monitoring™ Suite enables real-time response and team collaboration to share insights and track trends. 

Ready to begin your journey to more effective root cause analysis? Schedule a Seeq live demo today.