This site aims to complement the material published at:
Graafmans, T., Turetken, O., Poppelaars, H. et al. (2021) Process Mining for Six Sigma. Business Information Systems Engineering. 63, 277–300 (2021). https://doi.org/10.1007/s12599-020-00649-w

The literature acknowledges the applicability of process mining techniques in the DMAIC cycle of Six Sigma programs. However, the existing literature lacks a method or a guideline that is explicitly dedicated to how these techniques can be systematically used in the DMAIC cycle. Process Mining for Six Sigma (PMSS) is a guideline that shows how DMAIC activities of the Six Sigma initiatives can be supported with process mining techniques, thereby bridging the gap between Six Sigma and the field of process mining. The goal is to show how and where Process Mining techniques can be used to support the DMAIC of Six Sigma and potentially make it more efficient and effective to perform related DMAIC activities.

A graphical representation of PMSS is shown below.

Below we elaborate on the steps at each phase.

Define Phase

The Define phase incorporates three steps. In the Planning, the business goals and questions are identified, the processes that will be analyzed and improved are selected, and a project team is established. In order to help identify appropriate business goals and accompanying business questions (that related to one or more aspects of business processes, i.e., quality, time, resource, costs), the preliminary data preparation is performed, such that a brief overview of the process can be gathered on the event data in the exploratory mining & analysis step. However, before considering the event data, it is important to have clear business goals and questions that are expected to be addressed in the following steps.

The input for the planning step constitutes a broad spectrum of information regarding the organization, such as its business processes and major issues faced regarding these processes, a selection of processes that are known to possess issues and require attention for improvement, and the first insights gained from exploratory mining and analysis. As a result of a set of activities of this step the business goals and questions are defined, processes to be improved and supporting systems are identified, the project team with members that bring different perspectives to the process execution is composed, and a preliminary description of the business case is defined. The leading role in this step is the business user acting as the project owner and contributing with the domain knowledge and relevant context information to ensure that the initiative starts with the right direction. Although these activities in the planning step are depicted in an ideal sequence, they are iterative as their actual execution unfolds.

The objective in the preliminary data preparation step is to provide data for the coming exploratory data analysis, and in turn to facilitate the planning step in better identifying business problems and building a business case for the initiative. This step is a special form of data preparation and comprised of three sub-steps: (preliminary) data extraction, where the process execution data is extracted from selected information systems; processing, where the data is prepared so that it can be stored in an event log and can be loaded into a process mining tool; and verification, where the data reviewed for correctness and validity to ensure that it correct depicts the actual process. In the data processing sub-step, the good practices include the definition of an audit trail of the changes made to the data and the description of the data types in the log. At this step, the leading role is the data analyst.

The exploratory mining & analysis step is a special form of mining & analysis that is carried out with exploratory motives to support the identification of process-related business problems. It takes the data originating from the previous step as input and incorporates three families of techniques: process discovery, conformance checking and process analysis (covering also enhancement techniques). In addition to process mining techniques, the process analysis incorporates statistical methods and techniques (such as Pareto analysis, histograms, descriptive statistics. Such techniques are often used in Six Sigma initiatives to support traditional process analysis. The importance of traditional process analysis alongside process mining is stressed by multiple studies in the process mining field in which process analysis makes up for a large share of the actual reasoning. Note that the mining & analysis does not enforce any predefined order for the use of process discovery, conformance checking, and process analysis techniques.

The feedback arrow from the exploratory mining & analysis step to planning indicates that the insights gained from the exploratory analysis can serve as input for establishing the business goals. The process analyst ensures that the activities in this step are properly carried out.

The three steps in the Define phase, planning, preliminary data preparation, and exploratory mining & analysis can be conducted iteratively until the business goals for the improvement project are clearly defined. The define phase also clarifies additional execution-related information that should be extracted from the information systems the Measure phase.

Measure Phase

With the output of required additional information to be extracted from information systems, in the data preparation step of the Measure phase, relevant data and metrics are retrieved, a baseline is established, and current process performance is determined. This is achieved through the sub-steps of data extraction, processing and verification. The objective is to extract additional process execution-related data from the information systems of the organization, to process it to obtain a clear, filtered, and enriched event log, and to verify that the data is correct and valid and can be used as input to the next step, i.e., the explanatory mining & analysis. The leading role for this step is the data analyst.

Analyze Phase

In the Analyze phase, a closer look is taken at the data through explanatory mining & analysis. In this phase, the team guided by the process analyst performs a detailed analysis of the data with the aim to detect potential causes for the problems identified in the previous phase (Define), and identify improvement opportunities that can be acted upon in the next phase (Improve).

The purple outline for this step indicates that it comprises process discovery, conformance checking, and process analysis (as indicated in the legend by the block titled mining and analysis in the Figure). Although the techniques used in this phase are the same as those used in the exploratory mining & analysis step, the analysis is driven by identified business questions aiming to address the improvement goals. In turn, the level of analysis at this step is deeper and more intense. The main input constitutes the event logs created in the data preparation step, the audit trail, the data description, and business problems.

Improve Phase

The opportunities identified in the previous step are addressed at the Improvement phase. At this phase, process mining can be useful in detecting the likely impact of alternative improvement actions and in selecting those that are likely to bring the highest impact when implemented.

As the actual implementation of the process improvement action takes place in the business side, the business user takes the leading role in this step and implements and manages the required changes in the processes taking the improvement opportunities found in the previous step as input. The deliverable of this step issubsequently improvements or so-called process changes towards business goals.

Control Phase

In the Control phase, the process mining techniques can be used to monitor the predefined process performance indicators to help evaluate if the implemented changes yielded the expected results. In this phase, we can distinguish two steps: monitoring, where the process execution is monitored with respect to the performance indicators, and evaluation, where the organization determines if the impact of the changes depicts the expected results and whether the process has been successfully improved. As a result of monitoring, new findings may also emerge leading to new business problems and in turn a re-initiation of the DMAIC cycle. As we mentioned above, although Figure3 depicts a single feedback loop from the Control to the Define phase, PMSS assumes iterations that allow traversing between any prior phase.

In the monitoring step, the business user (e.g., the change manager) is the leading role to identify and prioritize the process performance indicators (and metrics) that are of importance to the organization and thus should be monitored. In addition, since it is important to have domain knowledge at this point and since the evaluation happens on the business side, the business user (e.g., improvement project leader) is responsible for evaluating the values of the indicators that are monitored in order to assess if the process improvements yielded the expected result.

Table 1 below provides more details regarding each PMSS step. Table 2 provides more details regarding each phase.