Process Discovery and Conformance Checking with Process Mining to Enable RPA Strategy
- July 17, 2019
- 0 Comments
[*Editor’s note: This is a guest post from Jacopo Cecchi, VP os Marketing at myInvenio. With offices in Boston (USA) and Reggio Emilia (Italy), myInvenio was recognized as Cool Vendor in Analytics by Gartner (2016) and as Hot Vendor in BPM by Aragon Research (2018). myInvenio is a visionary leader in process mining, leading innovation in the field with its strong commitment to R&D that benefits from active collaborations with prestigious academic institutions and OT Consulting, myInvenio’s consulting group. World class organizations trust myInvenio, the list includes: Capgemini, Chiesi Pharmaceuticals, CREDEM Bank, Deloitte, Fiat Chrysler, KPMG, Intesa San Paolo Bank, McKinsey & Co. For more information visit them at www.my-invenio.com]
Business processes often involve multiple stakeholders performing different parts of the process and relying on multiple applications and siloed information systems to run them. This fragmentation of people and information systems involved translates into a lack of visibility on the end-to-end process, which ultimately makes it difficult to understand how processes actually run and to check whether they are in compliance with their expected behavior.
Traditional approaches to business process design and modelling rely on qualitative and subjective data – such as interviews and questionnaires – to map and analyze current state business processes and involve, often, expensive consulting services. Process audit common practices also rely on limited data – e.g. audit sampling – to assess process compliance. As a result, traditional approaches to process discovery and conformance checking are suboptimal and expensive.
Process mining provides a new bottom-up, transactional data-driven approach to business process discovery and conformance checking that overcomes the limits of the aforementioned traditional approaches. Mining the event logs data stored in the information systems used by the process’ stakeholders, process mining allows to automatically generate a dynamic model of the As-Is process that is objective and comprehensive of all process variants. Not only the model has no observer’s bias, it also takes less time to be created and is less expensive.
The more advanced the process mining solution used, the more accurate the dynamic model of the process that can be achieved and the analysis that can be performed.
myInvenio leads innovation in the field, being able to mine process decision rules – e.g. in an accounts payable process myInvenio can infer automatically that above a certain amount an invoice requires some approval activities – and being able to map processes with many to many relationships in a single comprehensive process. The latter capability, referred to as Multi-Level Process Mining, allows to analyze complex processes such as Procure to Pay or Order to Cash treating them as single end-to-end processes without having to deal with biased statistics, data divergence and convergence issues. Without Multi-Level Process Mining, P2P subprocesses – purchasing, ordering, invoicing, payment – would have to be analyzed independently, losing visibility on the end-to-end P2P process.
The As-Is process model discovered with mining enables to analyze the process along different dimensions, which include control flow, time, costs, resources involved, and process instances. These analyses are a big support to process improvement initiatives, providing unprecedented fine grained visibility on the end-to-end process for its assessment.
A second important application of process mining is to monitor the conformance of the actual process against its reference model, for example the model developed on BPM solutions such as iGrafx. This capability of automatically identifying deviations between how the process is supposed to work and how it actually works empowers business and regulatory compliance monitoring, especially when compliance monitoring is implemented in real time – with all the benefits that derive from spotting deviations as soon as they occur.
Given the aforementioned discovery and conformance checking capabilities that process mining brings to the table, it is easy to understand how process mining enables and accelerates an RPA strategy:
- Before implementing RPA, process mining helps to discover the process, uncover redundancies and inefficiencies, detect non-compliant activities and process path variants; identify RPA’ s best candidate activities and support the standardization of process path variants. Few solutions like myInvenio go even beyond current state process assessment, providing the capability to simulate RPA’s impact on process behavior and ROI by performing accurate what-if scenarios analyses that minimize adoption risks.
- Once RPA has been deployed, process mining helps to monitor process performance and RPA’s ROI in real time and to ensure that RPA is working as expected. In this phase, process mining plays a key role also in automatically generating process and RPA documentation that can be used in repositories such as iGrafx’ for governance, risk and compliance purposes.
Facing increasing customers’ demand for tools to enable RPA strategy and to monitor process and RPA’s compliance, myInvenio and iGrafx are developing integration between their two solutions to seamlessly integrate mined data into iGrafx and augment iGrafx’ GRC capabilities. These capabilities are now available to the public as part of iGrafx’ RPA Accelerator offering, whose description can be found here.