A little over a year ago, in its 2022 PEAK Matrix Assessment of the process mining space, leading research group Everest noted that among the organizations it polled “Process simulation as a capability has very high interest.”
Simulation analysis in process mining shares a conceptual heritage with more familiar simulation use cases in engineering or physical product design. But it has far greater scope. Rather than using simulation modeling to refine the physical performance of, say, an engine, simulation analysis in process mining has the potential to refine the performance of entire businesses.
Simulation analysis provides insight into the likely consequences of process changes — across functions, people, products, systems and customers. It helps derisk decision making, providing a fact-based digital safe space in which to game out strategic business change across outcomes and processes.
In this article, we take a closer look at simulation analysis, how it works, why it’s important, and how it can be applied.
Simulation analysis is a process mining technique that allows businesses to look before they leap into process change. It enables them to model and understand potential outcomes before committing to any action.
Firstly, it requires the business to digitally model as-is processes and their interconnectivity. With this knowledge, companies can create digital simulations of how these processes might respond to any shifts in operational parameters — and the likely impact on business performance.
Simply put, simulation analysis enables an organization to test “what would happen if we changed XYZ processes?” and calculate likely scenarios using actual operational data as a basis. It also enables them to identify the optimum changes required for generating target outcomes.
For example, it could help explore the potential outcomes of automating manual processes, or shifting from a single supplier to multiple vendors. It could also help pinpoint the method(s) of new process execution with the minimum disruption and maximum beneficial outcomes.
The kind of insight from simulation analysis described above is why simulation analysis in process mining offers businesses such an important step forward.
Businesses are complex entities, comprising multiple overlapping, interacting, and interdependent functions, systems, people and processes. Consequently, any process change carries with it the potential for unintended consequences. A change in one process or parameter can ripple across a business, even where there isn’t an immediately visible connection between the two. The bigger the potential change, the more important it is to understand likely outcomes.
Using simulation analysis in process mining, businesses are able to make better decisions and maximize the output of strategic change initiatives. Take the case of Celonis customer iFood — a leading food ordering and food delivery platform in Brazil — for example. They’re looking to harness simulation analysis to help them achieve a successful scale-up.
“We are looking to go to process simulation because we believe that's the way we can scale,” said Karina Brumatti, Process Coordinator at iFood. “To grow, we have to understand where our bottlenecks are and simulating a process is the way to be proactive.”
By offering real insight into the impact of multiple changes to a process or processes — like a major scale-up — simulation analysis enables continuous performance improvement.
Finally, simulation analysis is a logical enhancement to an organization’s current process mining activity — a means to extend and amplify the value of the value of such activity. Let’s step through some key process mining stages to illustrate the point.
As part of the process discovery phase, historical event data from business systems is analyzed to create a process model or models. These are visual representations of real-world process flow, as they actually operate.
With such process models in hand, businesses can then carry out their conformance checking — comparing the actual processes to the expected or ideal process models. Conformance checking helps identify deviations between the actual process execution and what was expected.
With both the discovered process model and insights from conformance checking, simulation analysis can be employed to test different scenarios and potential changes to rectify the non-conformances. Businesses can simulate how changes in processes might impact its performance, resource utilization, and other factors. This helps in making informed decisions about process improvements.
But beyond that, simulation analysis can also be used to test whether the ‘ideal’ process models still represent the most effective means to achieve optimum business performance.
There are varying approaches to simulation analysis in process mining. A typical approach, however, comes in two parts:
Creating a digital visualization (or process digital twin) of as-is processes to act as a starting point or control from which to measure the impact of any simulated process changes.
Using statistical modeling — or more recently AI / machine learning-based modeling — to calculate and simulate the likely outcomes of process changes based on given parameters.
Let’s look at these individually.
It contains a Business Process Modeling Notation (BPMN) model extended with multiple statistical parameters about the process (such as the process impetus, the activities involved, resources required, processing times) but also about how each of these activities are connected, or impact each other — to create branching probabilities between them.
Historical data from event logs is used to show how key process parameters have responded to change to help calculate and simulate scenarios of how they might respond to process change in the future.
So, with the digital process twin as the starting point, organizations can test these different process change scenarios — and analyze their impact on business KPIs.
To be valuable, simulations need a credible level of accuracy in assigning likely outcomes, which is where probability calculations come into play.
This is a rapidly evolving space in which the use of generative AI and machine learning are starting to play a more significant role. In such instances, systems use AI algorithms that learn from process mining data to generate new process outcomes by mimicking the complex relationships between process parameters.
Another approach commonly used in simulation analysis (at least for now) is known as the Monte Carlo simulation.
This simulation models the probability of different outcomes in processes by repeatedly generating different inputs in specific process variables and calculating the corresponding outputs. By doing so multiple times, the Monte Carlo simulation creates a distribution of possible outcomes, allowing businesses to estimate probabilities and make informed decisions.
For example, let’s consider a hypothetical supply chain process where clothing orders are fulfilled from a distribution center (DC). The time taken to complete each step of the order from receipt to send-out can vary due to factors like worker efficiency, item availability and DC system downtime.
The Monte Carlo simulation enables us to understand the overall process performance by considering and building probability distributions for each activity's duration based on historical data. Rinse and repeat thousands of times and we arrive at a range of possible outcomes deriving from changes in any of the key processes — as well as identifying potential bottlenecks and opportunities for improvement.
The benefits of simulation analysis in process mining are clear. With process mining providing the data-based, factual underpinning for forward-looking simulation analysis, these two intersecting disciplines are perfect partners.
What’s truly exciting for business leaders, however, is the speed of development in both these areas. With innovations such as object-centric process mining (OCPM) and Celonis’ Process Sphere, organizations can now access end-to-end maps of their complex business processes and how they interact with each other. At the same time, the rapid development of sophisticated AI and machine learning tools looks set to offer ever-more accurate predictive analytics and simulation analysis. This means an ever clearer picture of the likely impact of process change on key business outputs.
Building out digital twin organizations (DTOs) sounds like science fiction. The complex interactions and processes across whole businesses accurately modeled and the impact of business change initiatives able to be simulated and analyzed? It’s early-ish days in an emerging field, but it’s not any kind of fiction, it’s happening.