HITL-AB-BPM: Business Process Improvement with AB Testing and Human-in-the-Loop

Aaron F. Kurz; Bennet Santelmann; Timo Großmann; Timotheus Kampik; Luise Pufahl; Ingo Weber

HITL-AB-BPM: Business Process Improvement with AB Testing and Human-in-the-Loop

Aaron F. Kurz, Bennet Santelmann, Timo Großmann, Timotheus Kampik, Luise Pufahl, Ingo Weber

Research output: Contribution to journal › Conference article › peer-review

1 Scopus citations

Abstract

Organizations improve their business processes regularly to react to clients’ requests, changing business environments, or regulations. However, applying a changed process design directly in an organization is risky: frequently, process changes do not result in actual improvements. In this demo, we present a tool for AB testing of process versions with reinforcement learning techniques and a new approach to human control of the procedure. The goal of the tool is a fair and fast comparison in the production environment. As the system might not have all information required to make informed decisions, we propose how human experts can be put in control.

Original language	English
Pages (from-to)	122-126
Number of pages	5
Journal	CEUR Workshop Proceedings
Volume	3216
State	Published - 2022
Externally published	Yes
Event	2022 Best Dissertation Award, Doctoral Consortium, and Demonstration and Resources Track at BPM, BPM-D 2022 - Munster, Germany Duration: 11 Sep 2022 → 16 Sep 2022

Keywords

AB testing
Human-in-the-loop
Multi-armed bandit
Process redesign

Cite this

@article{215b5a510c704c6cbbe816bbe9365dbf,

title = "HITL-AB-BPM: Business Process Improvement with AB Testing and Human-in-the-Loop",

abstract = "Organizations improve their business processes regularly to react to clients{\textquoteright} requests, changing business environments, or regulations. However, applying a changed process design directly in an organization is risky: frequently, process changes do not result in actual improvements. In this demo, we present a tool for AB testing of process versions with reinforcement learning techniques and a new approach to human control of the procedure. The goal of the tool is a fair and fast comparison in the production environment. As the system might not have all information required to make informed decisions, we propose how human experts can be put in control.",

keywords = "AB testing, Human-in-the-loop, Multi-armed bandit, Process redesign",

author = "Kurz, {Aaron F.} and Bennet Santelmann and Timo Gro{\ss}mann and Timotheus Kampik and Luise Pufahl and Ingo Weber",

note = "Publisher Copyright: {\textcopyright} 2022 Copyright for this paper by its authors.; 2022 Best Dissertation Award, Doctoral Consortium, and Demonstration and Resources Track at BPM, BPM-D 2022 ; Conference date: 11-09-2022 Through 16-09-2022",

year = "2022",

language = "English",

volume = "3216",

pages = "122--126",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

}

TY - JOUR

T1 - HITL-AB-BPM

T2 - 2022 Best Dissertation Award, Doctoral Consortium, and Demonstration and Resources Track at BPM, BPM-D 2022

AU - Kurz, Aaron F.

AU - Santelmann, Bennet

AU - Großmann, Timo

AU - Kampik, Timotheus

AU - Pufahl, Luise

AU - Weber, Ingo

PY - 2022

Y1 - 2022

N2 - Organizations improve their business processes regularly to react to clients’ requests, changing business environments, or regulations. However, applying a changed process design directly in an organization is risky: frequently, process changes do not result in actual improvements. In this demo, we present a tool for AB testing of process versions with reinforcement learning techniques and a new approach to human control of the procedure. The goal of the tool is a fair and fast comparison in the production environment. As the system might not have all information required to make informed decisions, we propose how human experts can be put in control.

AB - Organizations improve their business processes regularly to react to clients’ requests, changing business environments, or regulations. However, applying a changed process design directly in an organization is risky: frequently, process changes do not result in actual improvements. In this demo, we present a tool for AB testing of process versions with reinforcement learning techniques and a new approach to human control of the procedure. The goal of the tool is a fair and fast comparison in the production environment. As the system might not have all information required to make informed decisions, we propose how human experts can be put in control.

KW - AB testing

KW - Human-in-the-loop

KW - Multi-armed bandit

KW - Process redesign

UR - http://www.scopus.com/inward/record.url?scp=85139386488&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85139386488

SN - 1613-0073

VL - 3216

SP - 122

EP - 126

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

Y2 - 11 September 2022 through 16 September 2022

ER -

HITL-AB-BPM: Business Process Improvement with AB Testing and Human-in-the-Loop

Abstract

Keywords

Other files and links

Fingerprint

Cite this