Analyzing business processes with data

With digital transformation, business process management should be based on data. This can already be the case with traditional IT systems that embody business processes. But here, it is not about machine learning applied to data, but about analytics applied to logs data in order to reveal the real orchestration of business operations. The purpose of machine learning is to automate a task that is only a node in a business process, which is itself represented by a graph connecting several nodes.

First, though, we need to be sure of what those business processes are – and that is not always a straightforward proposition.

Why not? Because for each process, people have a subjective and partial view, depending on their individual roles and perspectives. What’s more, a process can be subject to constant change, making it difficult for its documentation to keep in step with its development. In fact, data scientists discover huge gaps between the business processes described by people working in and those revealed by process mining.

Process mining is a fast growing field in which data scientists apply analytics on data extracted from IT systems, namely event logs. These event logs are generated by all the components of an IT system that embodies a business process.

A plug and play process mining solution can be applied to the event logs that will reveal a full and accurate picture of the business process – not the process described in the manual, nor the process as perceived either by management or by individual front-line staff, but the actual process, with all its secret add-ons, workarounds, shortcuts, dead-ends, and compromises.

The main outcome of process mining is a graphical representation of the business process – a graph with interconnected nodes, each one corresponding to a task, as shown in this diagram (see Figure 4). The reality is of course more complex than any ideal, but it provides a sound and accurate base from which data scientists can redesign and optimize processes.

Figure 4. The ideal process vs. the process reality