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, 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 NTNU on the 2025 BRAGE paper.

Research themes

Predictive process monitoring — earliness, loss design, evaluation rigour

Focuses on remaining-cycle-time regression with LSTM-based architectures following the Tax et al. 2017 / 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

See also

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