---
title: "Mike Riess"
type: entity
tags: [person, author, ppm, prescriptive-process-monitoring, simulation, concept-drift, llm, telenor, nmbu, ntnu]
sources:
  - "[[sources/2023-riess-phd-thesis-ppm]]"
  - "[[sources/2022-riess-metaheuristics-concept-drift-survey]]"
  - "[[sources/2023-riess-temporal-loss-remaining-cycle-time]]"
  - "[[sources/2024-riess-synbps-simulation-framework]]"
  - "[[sources/2025-riess-jorgensen-brage-benchmark-norwegian-llm]]"
created: 2026-04-20
updated: 2026-04-20
---

# Mike Riess

Researcher in business process management, predictive & prescriptive process monitoring, concept drift, simulation, and (more recently) large-language-model benchmarking. Wiki owner; his publications are primary sources in this knowledge base.

## Bio
- **PhD** in Economics and Business, [[entities/nmbu|Norwegian University of Life Sciences (NMBU)]], School of Economics and Business, 2022 (published 2023, thesis number 2022:21). Main supervisor: Joachim Scholderer. Thesis: *Essays on Predictive and Prescriptive Process Monitoring* ([[sources/2023-riess-phd-thesis-ppm]]).
- **MSc** in Business Intelligence, Aarhus University, Denmark.
- **BSc** in Business Administration, Aarhus University, Denmark.
- ORCID: 0000-0002-4850-7121.
- Email: mike@riess.no (personal) / mike.riess@telenor.com (work).

## Affiliations timeline
- **?–2022 · NMBU, School of Economics and Business** — PhD researcher (Ås, Norway). Primary venue for Papers I–IV of the thesis.
- **2022–present · Telenor Group, Research and Innovation** — Senior research scientist (Oslo). Secondary academic affiliation with NMBU continued through 2024 publications; collaboration with [[entities/ntnu|NTNU]] on the 2025 BRAGE paper.

## Research themes

### Predictive process monitoring — earliness, loss design, evaluation rigour
Focuses on remaining-cycle-time regression with [[concepts/lstm-ppm|LSTM]]-based architectures following the [[sources/2017-tax-lstm-process-prediction|Tax et al. 2017]] / [[sources/2017-navarin-lstm-data-aware-remaining-time|Navarin et al. 2017]] lineage. Introduces **earliness** as a first-class model objective (not merely an evaluation aspect) via temporally weighted L1 losses, and **temporal consistency** (TC) — monotonic decrease of predictions — as a third evaluation axis beyond accuracy and earliness.
→ [[sources/2023-riess-temporal-loss-remaining-cycle-time]]

### Business process simulation for controlled evaluation
Critiques PPM's reliance on small public-event-log benchmarks; proposes **SynBPS**, a parametric simulation framework enabling independent manipulation of process memory (Markov order), state-space size, transition entropy, activity-duration distributions, case arrival rate, and process stability.
→ [[sources/2024-riess-synbps-simulation-framework]]

### Concept drift and model maintenance
Surveys the metaheuristics-for-drift-adaptation literature; documents the shift from single-task AutoML (feature selection, HPO) to Full Model Selection in online/streaming settings. Frames drift as a cost-vs-performance trade-off in ML-lifecycle management.
→ [[sources/2022-riess-metaheuristics-concept-drift-survey]]

### Prescriptive process monitoring — customer loyalty and queueing
Paper III of the thesis (with Scholderer) proposes predicted-loyalty queue prioritisation in customer service; evaluated via agent-based simulation calibrated on telecom data. Submitted to Decision Support Systems. *Referenced but not ingested as stand-alone source.*
→ [[sources/2023-riess-phd-thesis-ppm]] (Paper III)

### LLM benchmarking in low-resource Scandinavian contexts
Introduces **BRAGE**, a private benchmark for zero-shot LLM classification of Norwegian customer-service dialogues. Connects to the drift-adaptation motivation: zero-shot LLM classification is proposed as a lower-maintenance alternative to supervised classifiers that require re-training as call-topic distributions shift.
→ [[sources/2025-riess-jorgensen-brage-benchmark-norwegian-llm]]

## Publications in this wiki

| Year | Title | Venue | Role | Slug |
|---|---|---|---|---|
| 2022 | Automating model management: a survey on metaheuristics for concept-drift adaptation | J. Data, Information and Management 4:211–229 | Sole author | [[sources/2022-riess-metaheuristics-concept-drift-survey]] |
| 2023 | Remaining cycle time prediction: Temporal loss functions and prediction consistency | Nordic Machine Intelligence 3:12–26 | Sole author | [[sources/2023-riess-temporal-loss-remaining-cycle-time]] |
| 2023 | Essays on Predictive and Prescriptive Process Monitoring (PhD thesis) | NMBU Thesis 2022:21 | Sole author | [[sources/2023-riess-phd-thesis-ppm]] |
| 2024 | SynBPS: a parametric simulation framework for the generation of event-log data | SIMULATION (SCS) | Sole author | [[sources/2024-riess-synbps-simulation-framework]] |
| 2025 | The BRAGE Benchmark: Evaluating Zero-shot Learning Capabilities of LLMs for Norwegian Customer Service Dialogues | NoDaLiDa/Baltic-HLT 2025 | First author (w/ Jørgensen) | [[sources/2025-riess-jorgensen-brage-benchmark-norwegian-llm]] |

## Software
- **SynBPS** — open-source parametric BPS framework. PyPI: `pip install SynBPS`; GitHub: [mikeriess/SynBPS](https://github.com/mikeriess/SynBPS); demonstration: [mikeriess/SBPS_results](https://github.com/mikeriess/SBPS_results).
- **BRAGE** — code on [tnresearch/brage](https://github.com/tnresearch/brage) (benchmark data private; aggregated results public).

## See also
- [[syntheses/riess-research-arc]] — critical synthesis of Riess's research line.
- [[entities/nmbu]] · [[entities/telenor]] · [[entities/ntnu]]