---
title: "Predicting Process Behavior in WoMan"
type: source
tags: [ppm, symbolic, workflow, woman-framework]
authors: [Ferilli Stefano; Esposito Floriana; Redavid Domenico; Angelastro Sergio]
year: 2016
venue: "AI*IA 2016, LNAI 10037, Springer: 308–320"
kind: paper
raw_path: "raw/Predictive process monitoring/Predicting Process Behavior in WoMan ferilli2016.pdf"
created: 2026-04-13
updated: 2026-04-13
key_claims:
  - Process models learned via process mining can support prediction, not just conformance checking.
  - The WoMan workflow framework handles more expressive processes than Petri/Workflow nets; prediction methods can leverage this expressivity.
  - Two approaches — within-model next-activity prediction and candidate-model disambiguation — are proposed.
---

# Ferilli et al. 2016 — Predicting Process Behaviour in WoMan

Symbolic, logic-based PPM grounded in the **WoMan workflow framework** (University of Bari). Pre-deep-learning. Uses the expressive workflow representation of WoMan to:
1. Predict next activity within a known model.
2. Disambiguate which of several candidate models is being enacted.

## Position
Represents the **symbolic / inductive-logic-programming lineage** of PPM, parallel to the statistical / neural mainstream. Cited less frequently in neural PPM benchmarks but relevant for the interpretability-oriented strand of the field.

## Connections
**Concepts:** [[concepts/next-activity-prediction]] · [[concepts/predictive-process-monitoring]]