Dirk Fahland

I am trying to sketch the landscape of describing, analyzing, and managing processes outside the well-established paradigm of a “BPMN process” where a process is executed in instances, and each instance is completely isolated from all other instances.

Thinking about Processes

Let me introduce the term “process thinking”.

Process-thinking is the fundamental paradigm for understanding, designing, and implementing goal-oriented behaviors in social and technical systems and organizations of all kinds and sizes.

Process thinking structures the information flow between various actors and resources in terms of processes: several coherent steps designed to achieve common and individual goals together.

Throughout a process, multiple actors, resources, physical objects and information entities interact and synchronize with each other.

The scope of process thinking varies depending on the system and dynamics considered based on “how many dynamics to consider?” (outer scope) and “how many entities describe these dynamics?” (inner scope).

“BPMN Process Thinking” and Classical Process Mining
One Execution – Single Entity

BPMN and classical process mining focus primarily on describing and analyzing information handling dynamics as they are found in many administrative procedures, for instance in insurance companies or universities.

Processes are scoped in terms of individual cases (or documents) whose information is processed along a single process description independent of other cases, often in a workflow system. In terms of scoping, such processes encompass a single-dimensional inner scope (information processing) structured into a single-dimensional outer scope (along a single case).

One Execution – Multiple Entities

Most organizations operate multiple processes sharing data or materials which requires to consider multiple processes and objects and their interlinked dynamics together.

Process thinking around dynamics in manufacturing and retail organizations, such as Order-to-Cash or Purchase-to-Pay processes, is often supported by complex Enterprise Resource Planning (ERP) or Customer Relations Management (CRM) systems.

Processes here are centered around updating and managing a collection of shared and interlinked documents by various actors together leading to mutually dependent and interconnected dynamics of multiple objects and processes (multi-dimensional outer scope) with a focus on information processing (single dimensional inner scope).

Taking the system into the picture

While information processing is the dominant behavior analyzed in process mining, the dynamics of a process may also be characterized and analyzed in other dimensions.

For example, how actors and resources, physical materials, and the underlying systems participate in the processing of cases of the same process (inner scope of process thinking)

• How are actors and resources involved in the dynamics – and how does the involvement of actors and resources influence the dynamics, for instance through availability, workload, and capabilities?
• How are physical materials involved in the dynamics, for instance through transporting or storing large amounts of materials via conveyor belts or vehicles. How do their physical properties and constraints influences the dynamics?
• How are the underlying systems are involved in the dynamics, and how do their capabilities and limitations influence the dynamics, for instance through queueing, prioritizing, and assigning of work or the (reliability of) automation of steps?

the progress of a case depends on availability of information, actors, and corresponding materials alike

In most processes, these different factors of processing are not independent but influence each other as the progress of a case depends on availability of information, actors, and corresponding materials alike, and is subject to limited availability of processing resources, and physical limitations of the supporting systems, which requires multiple dimensions to characterize a single dynamic (inner scope).

Multiple Executions – One Entity

Processes for manufacturing and logistics, such as baggage handling at airports combine information handling with material flows.

Physical items are processed along a logical process flow – and at the same time have to be moved around a physical environment of conveyor belts, carts, machines, and workers. Steadiness of flow is the central process objective.

In this characteristic, the processing of one material item depends not only on the logical process it has to go through but also on all other items that surround it: they together define whether work accumulates at a particular machine, work cannot be completed at the desired quality, or target deadlines are met. Did your bag reach the flight?

Call centers and hospitals are other examples where the processing of one case highly depends on what happens with other cases. A long waiting time in a queue can make a customer service contact go very differently. The quality and next steps in a medical treatment depend on how well the medical staff can focus on your case.

These phenomena cannot be observed, analyzed, and improved when studying each case in isolation.

Multiple Executions – Multiple Entities

More advanced logistics operations, such as warehouse automation and manufacturing systems, also consider material flows that are being merged together, through batch processing and manufacturing steps.

Analyzing and improving processes in such systems requires both a multi-dimensional inner scope and a multi-dimensional outer scope.

And now? Let’s talk…

What are your thoughts on this? Feel free to join and post a response here!