Enhancing Operational Excellence with Augmented Business Process Management

Recent years have brought a stream of exciting developments in the field of Business Process Management (BPM). We have seen a breathtaking uptake of business process automation technology, such as Robotic Process Automation (RPA). We have witnessed the rise of process mining, and promising evolutions in the areas of predictive process analytics and digital process twins.

In the eyes of a business analyst, each of these technologies offers compelling opportunities to enhance operational excellence. However, if we look at these technologies in isolation, it is easy to miss the bigger picture and the wider space of opportunities that these technologies open when used jointly rather than applied in individual projects or silos.

If we step back for a moment and take a broad-minded business analyst perspective, we will see a key driver behind these trends: The maturation of advanced analytics and AI technology have created a chain from data to insights, from insights to actions, and from actions to business value. In other words, we are witnessing the gestation of a new approach to BPM: An approach that leverages data analytics and AI to achieve continuous process improvement based on data. We call this approach Augmented BPM.

As we move deeper into the 2020s, we will see further steps in the direction of Augmented BPM. This article explores the trends driving the emergence of Augmented BPM and how organizations can start benefitting from these trends.

What is Augmented BPM?

Augmented BPM is an approach to manage business processes that relies on data analytics and AI to inform process improvement decisions both at design-time and at runtime.

Augmented BPM is more than the use of analytics and AI to execute individual tasks or to automate decisions (e.g., using a machine learning component to classify customer complaints). It is about using analytics and AI across-the-board to continuously analyze, adapt, re-design, and monitor business processes.

The Augmented BPM Pyramid

To better understand the scope of Augmented BPM, it is useful to conceptualize it as a pyramid of capabilities, as illustrated in Figure 1.

Figure 1. The Augmented BPM pyramid

At the lower layer, we find descriptive process mining (process mining for short). Process mining is a family of techniques to analyze business processes using datasets extracted from a multitude of enterprise systems, such as Customer Relationship Management (CRM) systems, Enterprise Resource Planning (ERP) system, and Talent Management Systems (TMS). These datasets are called event logs. An event log is a collection of records, each of which captures the execution of an activity (or a step within an activity) in the context of a business process.

Process mining encompasses a range of techniques, which can be classified into four capabilities:

• Automated process discovery. The ability to discover process models from data in order to put into evidence the main pathways and exceptions and to highlight wastes (e.g., rework, overprocessing).

• Conformance checking. The ability to detect deviations with respect to desired pathways, including violations of compliance rules (e.g., purchase orders without invoices) or deviations between the observed execution flows and normative pathways.

• Performance mining. The ability to link quantitative performance measures to elements of a process, such as linking SLA violations to bottlenecks, linking excessive costs or defects to rework loops, etc.

• Variant analysis. The ability to identify positive and negative deviance in a process by comparing how the process is performed for different subsets of cases (e.g., in different regions).

These capabilities allow business analysts to identify bottlenecks, rework, compliance violations, and other friction points to analyze the causes of these friction points, and their impact on key performance indicators (KPIs). Numerous organizations are using these capabilities to guide their continuous process improvement efforts. The annual survey of process mining case studies compiled by HSPI lists over 1000 successful case studies, across virtually all industry verticals, all the way from logistics and manufacturing to healthcare, government, telco, and financial services. For example, one of Australia’s largest banks is using process mining to drive a 360-degrees re-platforming of operations, to adapt to challenges brought up by changes in the regulatory environment combined with competition from fintech players.

While process mining is a valuable capability on its own, its long-term value comes from the fact that it opens the door to a wealth of other capabilities. Indeed, the very same datasets that organizations collect for process mining can be used to build predictive models capable of telling us what will happen in the future.

This brings us to the second layer of the Augmented BPM pyramid: predictive process mining. While descriptive process mining allows us to understand how a process has performed historically, predictive process mining allows us to forecast how a process will unfold in the future. Predictive process mining encompasses two capabilities:

• Predictive process monitoring. The ability to predict future states of a process. For example, in an Order-to-Cash (O2) process, a predictive process monitoring system can predict whether the products ordered by a customer will be dispatched on time or late. Typically, predictive process monitoring is implemented using machine learning techniques. We first train a predictive model on historical data, and then we apply it to a stream of events to produce a stream of predictions.

• Digital Process Twins (DPTs). The ability to predict the impact of a process change. For example, let’s consider an O2C process executed on an ERP system. By applying descriptive process mining, we may find that a bottleneck in the packaging step of the process is causing many delays. By using process mining and machine learning, we can build a DPT, which can act as a replica of the process, enabling business analysts to test their ideas prior to deployment. For example, a business analyst can then use a DPT to simulate what will happen if we add more resources to the packaging step of a warehouse management process. The DPT can estimate the impact of this and other possible changes on the percentage of late deliveries. This capability allows managers to estimate the ROI of process improvement efforts and to direct these efforts more effectively.

Predicting what's coming ahead in a process is informative. But predictions only create value when they are followed by action. This brings us to the third layer of the Augmented BPM pyramid: prescriptive process improvement. Prescriptive process improvement is about turning predictions into actions, optimally targeted and timed to improve the performance of a process with respect to one or more KPIs.

In this layer of the pyramid, the focus shifts from “process mining” to “process improvement”. Process mining focuses on discovering patterns from data and using these patterns to describe a process or to make predictions. In the third layer of the pyramid, patterns are secondary. Instead, we deal with actions.

Prescriptive process improvement encompasses two capabilities:

• Prescriptive process monitoring. The ability to recommend actions to optimize the performance of a process with respect to one or more KPIs, in real-time or near-real-time. For example, a prescriptive process monitoring system may detect that a shipment is likely to be delayed. It may then recommend contacting the customer(s) who ordered these products and offer them the option to dispatch their products in two batches to minimize the impact of the delay.

• Automated process improvement. The ability to recommend changes to a process to strike a tradeoff between competing KPIs, for example, lowering costs while minimizing defect rate and cycle times. An automated process improvement system may recommend to a process owner to change the allocation rules and work schedules of some resources, to alleviate certain bottlenecks that occur at the beginning of each week, or it may recommend performing additional verification steps for some types of purchase orders to prevent mishandled orders.

Recommendations such as the above ones are generated using a technology known as causal inference. In a nutshell, causal inference is a set of techniques to discover causal relations between actions and outcomes from historical data. Prescriptive process improvement leverages these relations to determine in which cases (and when) it is best to perform certain actions. Causal inference is not a new kid in the block. Companies like Uber, Facebook, Microsoft, Booking, and AirBnB, routinely use causal inference to determine how to target their targeting campaigns (e.g., discounts) for maximum ROI. So far, however, the success of this technology has been largely confined to the B2C world. We are only now seeing it flowing into the realm of operations optimization.

In prescriptive process improvement, the machine recommends possible actions to the human actors. The human actors decide whether to apply these recommendations or to ignore them. In other words, the interaction between the system and the human actors is one-way. What if the process improvement actions were the result of a conversation between human actors and the AI?

This brings us to the fourth layer: Augmented BPM. Augmented BPM goes beyond prescriptive process improvement in terms of the autonomy of the business process execution system and the richness of interactions between the machine and the human actors. Although Augmented BPM is still a nascent concept, we can already pinpoint two distinctive themes:

• Conversational process optimization. The ability to automatically detect situations where the performance of the process degrades, to explain to human actors (e.g., the process owner), the causes of this performance degradation, and to evaluate counteractions with a human actor. For example, a conversational process optimizer may detect that some types of shipments are often delayed. It may then suggest to the process owner that these shipments should be re-routed. The process owner may decide to offer to the affected customers a choice between the current route or a faster route (at an additional fee). The system then offers these options to customers who find themselves in similar situations in future. Over time, the system learns which options are most popular among customers for different locations.

• Adaptive self-driving processes. The ability for an automated system to determine the possible next actions in a process, to determine which action to take next, and to detect situations where an escalation to a human actor is required. For example, a system may determine the verifications that should be done when a purchase order is received, based on historical execution data. When the system detects a new type of purchase order that it has never seen before, it escalates to the human operator, who determines which verifications should be done for this new type of order. The system records the decision of the human operator and applies it when a purchase order of this type is received again.

In this upper layer of the pyramid, we shift from "process improvement" to “process management”. Indeed, augmented BPM is not only about discovering patterns, or generating process re-design recommendations. Augmented BPM is an approach to handle the entire BPM lifecycle.

How can analysts drive their organizations to benefit from Augmented BPM?

For many analysts, augmented BPM may seem too futuristic to deserve any immediate attention. However, the first two layers of the pyramid are already widely used by business analysts worldwide and across almost every industry vertical. Also, the technology behind the third layer of the pyramid is rapidly evolving and already used successfully in B2C applications. The benefits of climbing the Augmented BPM pyramid are considerable. Organizations that do not make their steps to climb the pyramid are likely to be left behind. The opportunity cost is too large to ignore.

Analysts who are pondering how to drive their organization’s journey along the pyramid might benefit from keeping in mind three important points along the way.

1. Lay the foundations, start climbing, keep climbing, don't hold off. Many managers postpone the adoption of process mining by stating “we don’t have the data”, or “our data is not good enough”. Yes, getting the data for process mining is often a challenge. But the benefits have been demonstrated repeatedly, in thousands of successful deployments. And getting the data to do process mining opens many doors. The data that is used today for process mining can be used tomorrow for predictive process monitoring or to build digital process twins. Once the obstacle of data collection and curation has been overcome, the possibilities are endless. Note that task mining provides an additional channel for collecting data— when the enterprise system does not allow us to do so.
2. Don't skip the layers. The lower layers of the Augmented BPM pyramid provide a foundation to derive business value from the upper layers. Organizations that wish to maximize the benefits of adopting the upper-layer capabilities need to master the lower layers.
3. Align strategically and build governance incrementally. Any process mining, predictive monitoring, or prescriptive process improvement initiative needs to be grounded on the strategic priorities of the organization. The capabilities in the augmented BPM pyramid should first and foremost be applied to business processes that matter to the organization. It is also important to adopt these technologies incrementally, one process at a time. Over time, a governance structure is needed to ensure that the technologies in the pyramid create value predictably and repeatably. But before getting there, it is important to have a few success stories internally, to gain executive support, and to keep this support by showing that every capability in the augmented BPM pyramid produces tangible value.

Author: Prof. Marlon Dumas, Co-Founder of Apromore and Professor at University of Tartu

Prof. Marlon Dumas is Partnerships Manager and Co-Founder of Apromore, and Professor of Information Systems at the University of Tartu. For the past two decades, Marlon has conducted research and delivered business process management and process mining consultancy. He’s executed R&D projects funded by the European Research Council, the Australian Research Council, the US Army Research Lab, and multinational companies including SAP, Microsoft, and Swedbank. His research has led to 200+ scientific publications, 10 US and EU patents, and a textbook used in 250+ universities. He’s contributed core components of the Apromore open-source product since 2011.