Process mining helps organisations discover and improve the performance of their processes, identifying bottlenecks and other areas of improvement
When implemented correctly, process mining reduces costs, improves service level agreements (SLAs) and empowers teams to solve inefficiencies quickly. It’s become an asset for Business Analysts and Operations Excellence Consultants, but crucially, it provides management with rich insights about the performance of their operation.
For those organisations who have successfully implemented process mining, creating and analysing current state processes is entirely automated. So, process discovery is no longer a time-consuming and resource intensive practice. But what about those organisations who haven’t yet seen the benefits of process mining?
If you employ people to execute business processes using IT systems, then you should be exploring process mining.
How does process mining work?
Process mining combines data science and process analytics to mine data from information systems. For example, Enterprise Resource Planning (ERP) or Customer Relationship Management (CRM) tools, create event logs with every transaction which provides an audit trail of processes. This shows what work is being done, when, and by who.
Process mining software then uses this information to create a process ‘model’. This allows the end-to-end process to be examined, showing the detailed steps taken and any variations. Built in machine learning models then help give insights into any root cause of deviations. For example, it might point out that every time a new customer needs proof of address, the process is slowed down. These models enable management to see if their processes are performing efficiently, or if they aren’t, they provide the information needed to optimise them.
Choosing the right places to apply process mining is important. Organisations who apply it to processes that have already been digitised, i.e. processes that use core IT systems, tend to see the best results. It provides an evidence-based view of how processes are performing and it’s an easy sell to senior management once they see where problems and opportunities lie.
Different Types of Process Mining
There are three basic types of process mining: discovery, conformance, and enhancement:
- Discovery uses event logs to create a model without outside influence. Under this technique, no previous process models exist to inform the development of the new process model. As such, this type of process mining is the most widely adopted.
- Conformance compares expectation with reality and aims to identify any deviations from the expected model.
- Enhancement is sometimes known as performance mining and is used to improve an existing process model. For example, the output of conformance checking prompts the identification bottlenecks, allowing managers to optimise the process.
Why is it so Important?
Lean Six Sigma is a method that relies on a collaborative team effort to improve performance by systematically removing waste and reducing variation. Lean Six Sigma has proved itself to be an effective methodology for reducing operating costs and increasing return-on-investment. However, identifying opportunities and measuring the effects of improvements has been difficult. Process mining helps by identifying and quantifying the inefficiencies in processes and showing how effective any changes have been. The use of these processes not only reduces costs, but it also drives more innovation, quality, and better customer retention.
Process mining’s success can be demonstrated by the experience of a large insurance company. A major source of inefficiency and cost for the company was their end-to-end claims process: from FNOL (First Notice of Loss) through to claims assessment to final claims payout. Process mining was used to understand how approximately 300,000 claims were routed, which steps in the process had the longest lead times, which the most variation and why. The company found that a combination of manual data processing, handling various documents and managing multiple hand-offs between third parties during the claim’s assessment was adding operating costs whilst also impacting the customer experience negatively. Within eight weeks, process mining provided a rich map of the end-to-end claims process, with insights on pain-points and number of opportunities for improvement. A combination of user training, process automation and process improvement initiatives followed. The result was an astounding 43% improvement in cycle times and over 1200 hours of time savings generated each month.
Challenges to overcome
But process mining is still a relatively new discipline, so there are still some challenges to overcome, including:
- Data quality – finding, merging and cleaning data is usually required to enable process mining. Data is likely to be scattered over various data sources. It is often incomplete or contains different labels or levels of granularity. Accounting for these differences is important to the information that a process model yields.
- Process changes – sometimes processes change as they are being analysed, resulting in the process model shifting.
Organisations that are striving to become digital businesses need to enhance the ability to investigate and analyse processes. The adoption of new automation technologies, such as RPA (robotic process automation), machine learning and NLP (natural language processing), has proven that business leaders want to invest in technologies that improve business performance. Process mining is another tool that organisations will increasingly lean on to achieve their business outcomes.
About the Author
Dan Johnson, Director of Automation, Future Workforce. Dan’s expertise in intelligent automation began in 2011 when he was part of the team delivering the first RPA projects into COOP Financial Services. He later joined Accenture as the Insurance Process Automation Lead for UK&I before leading an Automation and AI team at a UK Bank. Clients trust his exceptional experience to deliver their intelligent automation programmes.