Overview
Process mining recovers process models, bottlenecks, and organisational handoff networks from event logs; recent work adds LLM assistance and BPMN extensions for human–agent collaboration. Hand-off design literature stresses confidence thresholds, accountability, and escalation rules when AI enters hybrid workflows. This project hypothesises that AI adoption outcomes follow from what deployment does to interface junctions — points where format, authority, or fidelity changes between human and machine regimes — and proposes a three-axis diagnostic (sensing, modelling, amplification capacity) grounded in the Viable System Model and the Good Regulator condition. This project makes that diagnostic computational on public event logs, with falsifiable predictions about interventions.
Prerequisite
L172 Information, Energy and Intelligence (IEI), or equivalent preparation in information theory, maximum entropy, and information geometry.
FAQs
- What are the prerequisites?
L172 Information, Energy and Intelligence (IEI), or equivalent preparation in information theory, maximum entropy, and information geometry.
- What will I learn in this Project?
You will learn process mining (discovery, enhancement, organisational mining), how to build handoff and social networks from event logs, and how to operationalise cybernetic viability ideas (sensing, modelling, amplification capacity) as measurable quantities. You will simulate AI interventions on real public logs and test predictions about bottleneck relief versus silent bypass.
- What is the objective of the project?
(1) Select a public BPI Challenge log (or a large public collaboration log such as GitHub pull requests / open government tickets) and build process, handoff, and organisational networks. (2) Operationalise sensing / modelling / amplification capacities using estimated rates (queue entropy, exception frequency, rework cycles, time-to-escalation). Identify candidate judgement junctions. (3) Simulate AI interventions: automate a node versus insert an external approval / escalation gate; pre-register which moves should relieve bottlenecks versus create silent bypass. (4) Validate against held-out time periods or synthetic ground-truth overlays. Deliverables: reusable diagnostic notebook, case study, methods chapter aimed at InfoTop WP3.
- How does this fit into the bigger picture?
WP3 studies institutions adopting AI as evolving information topographies. A student cannot run a Bank of Italy fieldwork programme in one thesis, but can build the methodological instrument on open data. This project is the technical bridge from IEI / cybernetic theory to organisational diagnosis, and a natural collaboration point with the Accelerate / policy lab strand of the group.
