Business rule extraction using decision tree machine learning techniques - A case study into smart returnable transport items

Aug 4, 2026

Upload by Aurora

Decision support systems increasingly rely on machine learning to automate decisions, yet extracting meaningful value from large datasets while keeping justification transparent remains difficult. Rather than improving interpretability of black-box models, the paper advocates inherently interpretable models. It presents an approach using decision tree classification to extract business rules from IoT data to predict the asset status of Smart Returnable Transport Items, validated with two years of Netherlands case data.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Keywords

Introduction

[Available online at www.sciencedirect.com](Available online at www.sciencedirect.com) ScienceDirect Procedia Computer Science 220 (2023) 446–455 The 14th International Conference on Ambient Systems, Networks and Technologies (ANT) March 15-17, 2023, Leuven, Belgium Business rule extraction using decision tree machine learning techniques: A case study into smart returnable transport items Rob Bemthuis∗, Wei Wang, Maria-Eugenia Iacob, Paul Havinga University of Twente, Drienerlolaan 5, 7522 NB Enschede, the Netherlands Abstract Decision support systems are becoming increasingly sophisticated (e.g., being machine learning-based), attempting to automate decisions as much as possible. However, it remains challenging to extract meaningful value from large quantities of data while also maintaining transparency in seeking justification for the choices made. Instead of creating methods for increasing the interpretability of black box models, one way forward is to design models that are inherently interpretable in the first place. Rule-based methods can automate decisions with great transparency and accuracy, helping to ensure compliance with regulations and adherence to organizational guidelines. In this paper, we propose an approach that uses a decision tree machine learning classification technique for extracting business rules from IoT-generated data to predict the asset status of Smart Returnable Transport Items (SRTIs) . Wereport on an industrial case study that uses two years of historical data, obtained from an SRTI provider in the Netherlands, to predict the status of smart pallets. We compare the performance with the results obtained by using a support-vector machine (SVM) technique. Our experiments show that our solution is both accurate and flexible in terms of business rule elicitation. The obtained decision trees are human-interpretable, can easily be combined with other decision-making techniques, and provide a prediction accuracy marginally higher than an SVM technique. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer-review under responsibility of the scientific committee of the Conference Program Chairs Keywords: business rules; machine learning; decision tree; logistics; smart pallets; case study 1. Introduction Recent advancements in sensor networks, cyber-physical systems, and the ubiquity of Internet-of-Things (IoT) have significantly increased the collection of data [24] . Today’s information systems continuously monitor the physical environment and produce large quantities of data that are a great source for deriving information to support decisionmaking [2, 41] . However, although context-aware devices and wireless communication provide advanced services [29], there are still many types of unpredictable disruptions that affect our daily lives [9, 43] and business activities ∗ Corresponding author. Tel.: +31 534897009. E-mail address: [r.h.bemthuis@utwente.nl](r.h.bemthuis@utwente.nl) 1877-0509 © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer-review under responsibility of the scientific committee of the Conference Program Chairs 10.1016/j.procs.2023.03.057 Rob Bemthuis et al. / Procedia Computer Science 220 (2023) 446–455 447 (such as those in supply chains) that cannot be avoided. One of the causes is that extracting meaningful information from data is a complex and challenging task, requiring innovative techniques and algorithms to analyze and understand such data [46] . Yet, although the sophistication of data-driven artificial intelligence (AI) approaches has recently increased to such an extent that human intervention is minimized [6],

FAQs

Why is extracting business value and transparency from large ML datasets challenging?

Large datasets make it difficult to derive meaningful value while still providing understandable justification for model choices. Many ML models also behave as black boxes, reducing trust.

What approach does the paper propose for business rule extraction?

The paper proposes using a decision tree machine learning classification technique to extract business rules from IoT-generated data and predict the asset status of Smart Returnable Transport Items.

How does the decision tree method compare with SVM in the reported experiments?

The experiments show the decision trees are human-interpretable and flexible for eliciting business rules, with prediction accuracy marginally higher than the SVM approach.

Aug 4, 2026

Upload by Aurora