What is process mining?

Process mining can essentially be considered a cross between business process management (BPM) and data mining. Its aim is to gain an overview of the processes being used in an organization, enabling practitioners to optimize, enhance or swap out processes entirely.

In this guide we outline the benefits of process mining and how you can implement it at your organization.

Contents

• What is process mining?
• Event logs, models and metrics
• Types of process mining
• How can you implement process mining?
• Careers in process mining
• Real-world examples of the benefits of process mining
• The future of process mining

What is process mining?

Process mining is the technology used to analyze and optimize business processes. It enables the discovery of valuable insights, ensures transparency and drives continuous improvement with the objective of turning data gathered into actions that lead to greater efficiencies.

Many organizations will already be sitting on the data that they need in the form of event logs, and turning that data into business intelligence is process mining.

In discussion with PEX Network, Nina van Krimpen, business process consultant at chemicals company LyondellBasell, shared her definition of process mining: “Process mining provides end-to-end process transparency and visibility, helping organizations compare what they think is happening within their business with what is actually happening.

“The technology is also a great tool for global process owners to understand the level of standardization and maturity of end-to-end processes across a business. The aim of the technology is to have a tool that will show you where you are and what you need to do.”

Compared to standard reporting tools for processes such as business intelligence reports and dashboards, process mining offers insight into what organizations need to do to improve processes and drive them forward.

It is now being seen as an essential part of future proofing an organization in a world where technology is rapidly changing the landscape. Process mining helps organizations adapt quickly by continuously analyzing data and developing insights that drive new value, and plays a large part in guiding organizations toward successfully executing automation projects.

While it has proven to be beneficial in many scenarios, the use of process mining techniques is not without its limitations. Wil van der Aalst, known as the ‘godfather of process mining’, explained one of the primary limitations in conversation with PEX Network:

“Commercial process mining tools tend to focus on the discovery of directly-follows graphs based on event data, however, rather than using more sophisticated notations that are also able to capture concurrency. Although this is valuable, there remains a gap between process modeling tools and process mining tools.

"Organizations can get confused when seeing the real process models with all deviations and details while having oversimplified process models that only exist on paper. Here, conformance checking plays a pivotal role as it can bridge the gap between the high-level normative process models and the recorded event data.”

Event logs, models and metrics 

There are three main concepts that form the basis of process mining:

• Event logs: These are the foundation of process mining, capturing the sequential data of activities performed within a process. Event logs give a detailed view of the execution of a process.
• Process models: These are visual representations that show the relationship and flow between activities in a process.
• Process metrics: These help measure performance and efficiency, by quantifying things like cycle time, throughput time and bottleneck identification.

Types of process mining

There are four stages to successful process mining: collecting data, discovering what it says about processes, enhancing processes by removing friction points and finally monitoring the process to observe results. In more detail, these are: 

1. Process discovery: Using data from an event log, this technique creates a visualization of process workflows.
2. Conformance checking: This uses event data to identify deviations modelled behavior and actual execution behavior.
3. Process enhancement: This uses information discovered about a process to enhance it or extend a target model.
4. Operational support: This is an approach that involves influencing the process without completely reengineering it.

How can you implement process mining?

Olga Kacprzak, manager of process mining CoE at Poland’s fifth largest universal banking group mBank, advised practitioners to get to know their processes and the way they function before embarking on their process journey.

She notes that using Lean Six Sigma is effective and will help organizations know where to begin process optimization and how best to focus their efforts, while data is needed to gain the visibility needed over what processes look like in reality.

She offered three tips for organizations to successfully implement process mining:

1. Data science is crucial. Kacprzak stresses the importance of including data scientists to assist with data pre-processing. Involving the data science team should be a top priority in order to build accurate event logs.
   
   
2. Be patient. Successful process mining takes a lot of time and effort. Kacprzak advises practitioners to be patient, as big wins cannot be attained in just a few months.
   
   
3. Know your processes. Knowing how processes play out in reality will help organizations understand where they should start. Understanding where bottlenecks and inefficiencies exist allows practitioners to focus  efforts where they are needed most.

LyondellBasell’s Krimpen argues that global process owners should be brought in to support process mining projects: “To ensure success you need a good driver in the organization and support from top management that understands the value of the tool. Having a nice new dashboard that looks great will not change anything if there is no organizational structure to support it.”

That organizational structure means that the global process owner has ambassadors in functions across that process, Meanwhile the process mining team give visibility over the actions taking place and how they drive results.

Careers in process mining

There is a growing demand for skilled process mining professionals as adoption increases across different industries. Jobs vary depending on the organization, but the most common roles include analysts, engineers and data scientists. 

• Process mining analysts are responsible for conducting analyses using a variety of metrics and analytics tools, interpreting results and providing insights to stakeholders to help them improve their processes.
• Process mining engineers build and maintain data models and dashboards to carry out analysis. They develop software, design algorithms and validate these in line with the needs of the organization. 
• Data scientists in process mining apply advanced statistical and machine learning techniques to analyze process data and extract insights from it. They build visualizations and use modelling techniques to forecast process behavior and identify patterns, feeding this back to engineering teams.

Read our complete guide to careers in process mining, including where to study and the most typical jobs companies want to fill. 

Real-world examples from leading companies

How Sanofi is improving employee experience with process mining

Sanofi implemented process mining as part of its automation Center of Excellence (CoE), to analyze its operational processes and eliminate inefficiencies. Tiphaine Chatin, automation project manager officer at Sanofi, and Nikolay Goldovich, head of data, process, automation intelligence center of excellence at Sanofi Business Services, explained: “We boost it using business process model and notation (BPMN) to design core models and we call this whole framework process intelligence.”

“Process mining’s implementation is ongoing at Sanofi, with the aim being to derive KPI and reporting for performance transparency. We also want to ensure core model conformance to have Sanofi maturity level “as-is” as opposed to “to-be” and perform continuous improvement and development through fact-based ideas.”

They found various benefits through implementing process mining. It helped identify automation opportunities, core model compliance, audit and features
like workflow combination. “One of our best success stories is our job change simplification using the data from our human resources platform,” Chatin and Goldovich said.

“We wanted to provide a more intuitive way for managers to conduct actions. We reduced the number of steps they have to go through in the platform from six to two and they now need to complete 80 percent less data fields than before. We also added a virtual assistant to guide the users.”

Tackling bottlenecks at Intel Corporation

Steven Remsen, enterprise process excellence manager at Intel Corporation, identified a tool on the production line that was causing a bottleneck. He wanted to look at the downtime of the tool, i.e. when it was not processing work.

By modelling this cycle, mining the event logs to produce an accurate process model and comparing these models, Remsen was able to discover new complexities in what was generally considered to be a well-understood process.

The process was then enhanced through the application of machine learning to adjust the actions taken, which resulted in these bottlenecks were uplifted and this generated significant returns requiring minimal investment. Remsen noted that “this had a significant business impact for Intel, to the order of millions of dollars”.

Learn more about Intel Corporation's journey by watching Remsen's session below.

Driving process visibility at Neste

Markko Rajatora, vice-president of business processes at Neste, notes that through the implementation of process mining, Neste has been able to identify a number of problematic instances within processes, including duplicate deliveries and invoices, missing actions and late or missed deliveries. Prior to mining processes, the organization as a whole did not know how processes were operating on such an intricate and individual level.

“Now the functionality of our processes is clear and evident through the data we have gathered," Rajatora explains. "We can build KPIs and see metrics like lead times and sensitivities and take action to address details that break system limitations. It was process mining that showed us what was happening, and allowed us to dig into root causes and smooth out our cash-to-cash cycle throughout the Covid-19 pandemic.”

Learn more about Neste’s use of process mining for its cash-to-cash cycle during the Covid-19 pandemic by watching Rajatora’s session below.

The future of process mining  

Process mining was once considered an academic pursuit, though interest from business has grown since its inception. According to research featured on PEX Network, the global market is expected to reach $7.11bn by 2025. Meanwhile, the market size for software will expand at a compound annual growth rate of 50.1 percent between 2021 and 2028. 

The pandemic led to increased investment in process mining solutions, as organizations under serious pressure needed to develop a plan of action that ensured business continuity despite the disruptions.

Experts believe this is set to continue, especially given the advances in other analytical and data-related technology such as artificial intelligence. This opens up opportunities for process mining to be integrated with tools used for predictive analytics and data visualization, for example.

Want to learn more about process mining?

• Three steps to make process mining work for you - How to leverage process and task mining to overhaul processes and achieve operational excellence. 
• Data mining is coming to accelerate your continuous improvement – Discover how advances in data are helping process experts to analyze and optimize processes.
• Hybrid intelligence: To automate or not to automate – Prof.dr.ir Wil van der Aalst shares his thoughts on the concept of hybrid intelligence and the role of RPA. 
• Check out our annual event All Access Process Mining and watch sessions led by experts on demand. 

PEX Report 2024 global state of process excellence

PEX Network's annual report reveals the current state and future trends of process excellence. Based on a study profiling hundreds of global process excellence leaders, the report provides insights on how to leverage innovation, generative AI, and process intelligence to achieve operational excellence, customer-centricity, and organizational growth. Download your copy of the PEX Report 2024/2025 today and benchmark your process excellence journey against your peers.

Download Now

This article was originally written by Ian Hawkins on 8 January 2019 and was updated on 20 July 2023.