Structural Deviation Analysis Agent – Automatically find skipped steps, wrong sequences and unexpected paths in your business processes
Finds skipped steps, swapped sequences and paths that should not exist according to the target process – and shows what they cost.
What does the Structural Deviation Analysis Agent do?
The Structural Deviation Analysis Agent checks whether your processes run the way they are supposed to. It compares every case with the expected flow in Noreja’s causal process graph and finds structural deviations: skipped activities, swapped sequences, extra steps, loops and cases that end in the wrong place. For each deviation it reports how many cases are affected, how cycle time and effort change, and where the deviation clusters.
Questions the Structural Deviation Analysis Agent answers
“Which activities are skipped most often in the purchase-to-pay process?”
“In how many cases was the invoice posted before the goods receipt?”
“Which deviations from the expected flow cost us the most cycle time?”
“Show me the variants the target process does not allow for, sorted by case count.”
How the Structural Deviation Analysis Agent works
- 1
Read the target flow
The agent reads the expected flow of the perspective from the causal process graph: which activity should follow which, and which steps are mandatory. It also draws on process descriptions from the Context Hub.
- 2
Classify deviations
It assigns every deviation to a type: missing step, wrong sequence, extra step, repetition or premature end.
- 3
Assess the impact
It compares deviating and conforming cases: how much longer do they take, how many activities are added, what risks arise?
- 4
Narrow down the origin
Using case attributes – supplier, plant, document type, organizational unit – it narrows down where a deviation clusters.
KPIs and methods
- Conformance rate
- Share of cases that run through the process without a structural deviation.
- Deviations by type
- Missing, swapped and extra steps – counted separately.
- Affected cases and variants
- How many cases show a deviation and in which variants it occurs.
- Time impact
- Additional duration of deviating cases compared with conforming ones.
A practical example
In purchase-to-pay, the invoice should only be posted after the goods receipt. The agent finds the cases where that order is reversed, counts them, compares their cycle time with the remaining cases and shows that they concentrate on a few suppliers and one document type. You get a table of deviations ranked by frequency and added duration – and, if you like, a short presentation for the process owners.
Matching end-to-end processes
What you get
A list of structural deviations ranked by impact, with case counts, added duration and the attributes where they cluster. Every activity mentioned is linked to the process graph; tables can be exported to Excel.
What data the agent needs
A perspective in Noreja containing the process you want to check. The agent reads the expected flow from the causal process graph; process descriptions or policies in the Context Hub sharpen the picture.
Structural Deviation Analysis Agent: frequently asked questions
What is a structural deviation analysis of a process?
A structural deviation analysis checks the sequence and completeness of the steps in a business process. It shows where cases depart from the intended flow – because an approval is missing, steps are swapped or activities are repeated. In process analysis it is closely related to conformance checking.
How does the agent differ from classic conformance checking?
Classic conformance checking compares an event log with a modeled target process and returns a fitness score. The agent works on the causal process graph, where the dependencies between activities are already represented, and answers the follow-up questions too: which deviation costs the most time, and where does it cluster?
Do I need a BPMN model as the target process?
No. The expected flow is stored in Noreja’s process graph. If you have process descriptions or policies, put them in the Context Hub – the agent will use them as well.
Part of Minerva 2.0
The Structural Deviation Analysis Agent is an agent in Minerva, the AI assistant in Noreja. It uses the same data sources, the same choice of model – local or cloud – and the same access rights: it only reads what you are allowed to see and changes nothing in Noreja.
More agents from the library
Rework Analysis Agent
Calculates the rework rate, finds loops and repeated steps and connects them with their causes.
View agentCompliance Agent
Checks every case against your policies – approval limits, four-eyes principle, segregation of duties – and backs up each finding.
View agentRoot Cause Analysis Agent
Answers the question of why: traces dependencies in the causal process graph and explains problems with evidence.
View agentSee what the Structural Deviation Analysis Agent finds in your process data
In a demo we show you what answers the agent delivers on one of your processes – with numbers, reasoning and sources.
Book a demo