Variability (BPM 2016)

Modelling variability in process performance indicators

A business process (BP) may vary according to its specific context, due to changes in original process requirements, by the evolution of its environment of application, to reflect new allocation of responsibilities, new strategic and business goals, or by changes in general inputs of business process.
The modelling of business process variability (BPV) focuses on identifying variable and invariable parts of a business process (e.g., its control-flow, data or resources) with the aim of managing different versions of the same process together.

The performance perspective of business processes is concerned with the definition of performance requirements, usually as a set of Process Performance Indicators (PPIs), that address different dimensions like time, cost and quality. A PPI can be defined as a quanti able metric focused on evaluating the performance of a BP in terms of eficiency and eff ectiveness. They are measured directly by data generated within the process flow and are used for process controlling and continuous optimization.

Like other business process perspectives such as control-flow or data, there are cases in which PPIs are subject to variability. This variability can be related to variations that take place in other perspectives (e.g., if the activity the PPI measures does not appear in a certain process variant), but PPIs can also be subject to their own variations regardless of the other perspectives (e.g., the target value for a PPI in an incident management process may change depending on the criticality of the incident without any changes in the control-flow).

After a study of several business process variability cases and an analysis of different models to represent process performance indicators, we have identified
two dimensions of change in the de nition of PPIs, namely:

  • D1: A PPI varies depending on whether it is defined for all process variants or not.
  • D2: A PPI varies depending on attributes required to define it, which may change depending on the variant in which it is defined.

PPI variability is reflected according to D2, considering that a PPI varies if at least one of the following attributes changes: Target, Scope, Human resources or Measure definitions.

In the last case there are two dimensions of change, one related to the measure de nition itself and another one related to the relationship with the process:

  • D2.M1: A measure de nition maintains its structure, but may vary depending only on the business process elements to which it is connected.
  • D2.M2: A measure de nition changes its structure and may vary depending on the requirements of the process variant.

To support the modelling of the variability of PPIs, we extend the PPINOT Metamodel and the Visual PPINOT graphical notation, based on the metamodel.
For more information, see:

 

Examples of Variability in PPIs

SCOR is an example that reflects the dimensions of change related to PPI variability. The Supply Chain Operation Reference (SCOR) model is a framework that describes business activities related to all phases to supply customers demand. We focus on two elements of its structure: processes and measure de nitions (called metrics in SCOR), because due to its structure and the definition of its components, they have variability.

Deliver SCOR process can be implemented in four different ways depending on the selected strategy: D1-Deliver Stocked Product, D2-Deliver Make to Order Product, D3-Deliver Engineering to Order Product and D4-Deliver Retail Product. Each of them is a process variant (PV) of Deliver. They have a set of common tasks among them, but also have differences depending on the strategy selected.

We include two examples to show the variability on PPIs, based on the SCOR processes and measure definitions. Below, we show an example in which a SCOR measure definition has been used to specify a PPI. This measure is defined for three variants of the Deliver process. First, we use the formal definition of Visual PPINOT to represent the measure in each process variant. Then, we use the formal definition of the extension of Visual PPINOT.

 

 

In the second example we represent the variability in SCOR processes and measures by means of a graphical extension of Visual PPINOT.
For simplicity, we only focus on the three first process variants, because D-4 is totally different from other process variants. Those process variants are shown in the figure below as a C-EPC model. In it, 5 PPIs are defined on the basis of SCOR measure definitions as example of casuistry derived from variability dimensions proposed.

Deliver SCOR process modeled using C-EPC

Deliver SCOR process modeled using C-EPC (Larger image)

Those PPIs represent the dimensions of variability proposed.

  • RS.3.16 is applied for all variants without changes in it. (There is no variability for D1 neither D2)
  • RL.3.4 is also defined for all variants, but it is connected to different tasks depending on the process variant selected. The PPI does not depend on the BP control-flow. (Variability by D2)
  • RL.3.31 is defined for all PV, but depends on changes of the BP control-flow. Task D1.11 exists only in PV-1. (Variability by D2)
  • RS.3.100 also depends on changes of the BP control-flow, and it is only defined for PV-1. (Variability by D1 and D2)
  • RL.3.16 does not depend on the BP control-flow but it is defined only for PV-1 and PV-2 because of specific requirements for the PPI. (Variability by D1 and D2)
Top