In the business world, it is tautological to assert that good innovation, whether tangible or conceptual, one that reduces operational costs, increases productivity and improves transparency will receive widespread adoption and popularity. One such innovation is the concept of Business Process Management (BPM) which outlines a process-oriented approach to a business model that allows one to examine each distinct process and propose improvements within the model in a modular fashion.
Any good business model will invariably incorporate some concepts inherent to BPM. A creative management will concretely implement ideas outlined within the BPM framework in terms of offering flexibility and allowing for potential growth for the business. However, such attempts at streamlining the business model do not realize the full potential offered by BPM. Instead, by utilizing the guidelines provided by BPM to create a roadmap towards automating critical business processes, the creative mind is free to think of the big picture rather than get bogged down by detail. This important concept of Business Process Automation (BPA), or workflow management, allows for monitoring the performance of various business processes, identifying troubled tasks that require resource reallocation and specifying benchmarks for future performance.
In order to fully achieve BPA, all processes within the business model must be identified and described in detail. This description of the workflow within a process consists of three components: individual tasks to be performed and their execution order, flow of data and resource dependence. By encapsulating data as well as tasks to be performed using that data and available resources, one achieves modularity in a way that is inherently consistent and fault-tolerant. Furthermore, information about resources is available throughout the process, thus avoiding unnecessary verification each time a resource is required. By implementing these concepts, each business process achieves flexibility in terms of handling changes in resource availability and task specifications in a timely manner. Testing for changes before actual deployment by performing analysis and simulation can decide whether correctness criteria are met and desirable outcome is obtained. Problems within a process are detected early and effectively through monitoring, which will provide a natural way to cause escalation of tasks that demand attention. Post-processing of event logs, or process mining, can be used to suggest improvements. These steps that manage workflow--creation from anew or by repurposing an existing model, analytic and/or simulated testing, deployment, monitoring, process mining, improving--describe a BPM life-cycle for a process.
Depending on the complexity required and the underlying architecture, actual implementation or creation of a business process can be as simple as choosing an existing model or as involved as creating an information system and describing its details from scratch. This is followed by process enactment, where the process is executed, and process monitoring, where various performance indicators are monitored as well as logged for later use. Afterward, a business process may enter the diagnostic phase which draws upon information gathered from the monitoring phase to suggest improvements. The results from this phase may form a basis for suggesting adaptation to the process, thereby creating a new model ready for testing, and the cycle repeats. Allowing this life-cycle of a process to be automated lets a business adapt readily to changing circumstances.
Different methods have been prescribed for capturing business processes. Opinions vary on the applicability of these methods based on the need to ultimately analyze and automate processes. This lack of consensus may be attributable to several reasons such as inherent complexity and non-adherence to standards. It is possible that tasks within a process may have unusually complicated temporal or resource dependencies, pushing against the limit of ideas describable by process modeling languages. It is also possible that lack of common business standards for describing processes results in poor BPA implementations.
In the field of BPA, when describing business processes, one may notice similar sets of tasks, data flow and resource dependencies across various processes. This is a workflow pattern and it can be abstracted out of actual implementations and carefully analyzed on its own. This pattern may subsequently be used during process modeling for a variety of processes, thus making the pattern reusable.