The key distinction of Process Mining is its flexibility. In Part 2 of our two-part series, we discuss three specific use cases in digital transformation for telecom companies. 

Process Mining Use Cases for Telcos

The Role of Process Mining in Digital Transformation: Part 2

By Tunde Ibiyemi

In Part 1 of this series, we reviewed how traditional Business Process Management (BPM) techniques for process improvement are not always well suited to Telcos pursuing digital transformation. Firms that are seeking greater responsiveness in a more dynamic environment, need a more flexible approach to the development, deployment and management of processes. This is a key benefit of Process Mining. 

Process Mining is a nascent technique for process improvement that helps organizations harness a key by-product of digital transformation: vast amounts of data. It is an umbrella term for a collection of data mining and analytical techniques that are used to extract process insight from event logs generated by IT systems.

3 Use Cases for Process Mining 

Process mining can be used in three distinct ways.  

The first is Process Discovery, which is used when either there is no documented process model, the process has not been captured in a BPM system, or there is suspicion that the documented process is not a good reflection of its execution in reality.

7300-ProcessMining-2

The goal of process mining in this instance is to map the true process being executed, identify common and infrequent execution paths, bottlenecks, dependencies, and key resource constraints at each stage. 

The second, is Conformance Checking. Here, a company already has a process model, but would like to evaluate how closely the live execution of the process conforms to a documented model. This technique can be applied to a variety of model types including organizational models, business rules and procedures. 

7300-ProcessMining-3

The third is Process Enhancement, where the aim is to enrich an existing model by adding extra information about bottlenecks, deviations, throughput times, etc. 

7300-ProcessMining-4

In all three cases, the output can be provided in a variety of industry-standard notation, such as BPMN, EPC, and UML diagrams. Where an input process model is required, it can be provided in the same format. 

In addition to these broad use cases that are primarily conducted offline, process mining can be integrated with live systems to help with the prediction of process run time, identification of deviations or bottlenecks. These can be used to generate alerts or directly trigger corrective action. 

Process Mining can also be used to zoom in on the organization involvement in a process, eliciting information about who the key actors are, how and when they are involved in the process, and their interconnectedness. 

Where Process Mining Fits in Today

Process Mining techniques are not expected to supplant the BPM process. Rather, by making BPM data-driven, they will make it more responsive to operator needs today, and faster to identify and remedy process issues in the future. 

Notwithstanding these potential benefits, there are obstacles to Process Mining implementation. One being that many of the commercial tools are still in their infancy. The main providers are start-ups, including Celonis, Fluxicon, Minit and QPR Software. Some prominent BPM software providers (e.g. SAP and Software AG) are adding – and improving – the Process Mining capabilities of their existing offerings. 

Another impediment is the lack of understanding of Process Mining within most organizations. The skills and knowledge required for embedding Process Mining within BPM are not readily available within teams that would usually be tasked with delivery process improvement. 

Our recommendations to operators encountering these challenges is as follows:

  1. Identify a small number of processes to pilot Process Mining.
  2. Begin with a limited use case, possibly centring on process discovery.
  3. Use the pilot to evaluate and benchmark process mining against current practices.
  4. Build on the pilot to develop the expertise required to deliver and educate the rest of the organization on the advantages of Process Mining.

Conclusion

While Process Mining will not completely upend current BPM practices, it will significantly enhance them. Process Mining will make BPM faster to implement. It will provide insights to support decision-making prior to embarking on expensive transformation projects. And last but not least, it will make the realized benefits easier to identify and track.<>