Predicting supply chain risks using machine learning

The trade-off between performance and interpretability

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Managing supply chain risks has received increased attention in recent years, aiming to shield supply chains from disruptions by predicting their occurrence and mitigating their adverse effects. At the same time, the resurgence of Artificial Intelligence (AI) has led to the investigation of machine learning techniques and their applicability in supply chain risk management. However, most works focus on prediction performance and neglect the importance of interpretability so that results can be understood by supply chain practitioners, helping them make decisions that can mitigate or prevent risks from occurring. In this work, we first propose a supply chain risk prediction framework using data-driven AI techniques and relying on the synergy between AI and supply chain experts. We then explore the trade-off between prediction performance and interpretability by implementing and applying the framework on the case of predicting delivery delays in a real-world multi-tier manufacturing supply chain. Experiment results show that prioritising interpretability over performance may require a level of compromise, especially with regard to average precision scores.
Original languageEnglish
Pages (from-to)993-1004
Number of pages12
JournalFuture Generation Computer Systems
Volume101
Early online date29 Jul 2019
DOIs
Publication statusE-pub ahead of print - 29 Jul 2019

Fingerprint

Supply chains
Learning systems
Artificial intelligence
Supply chain management
Risk management
Experiments

Cite this

@article{b6e4fb8412f547678df15594508934c9,
title = "Predicting supply chain risks using machine learning: The trade-off between performance and interpretability",
abstract = "Managing supply chain risks has received increased attention in recent years, aiming to shield supply chains from disruptions by predicting their occurrence and mitigating their adverse effects. At the same time, the resurgence of Artificial Intelligence (AI) has led to the investigation of machine learning techniques and their applicability in supply chain risk management. However, most works focus on prediction performance and neglect the importance of interpretability so that results can be understood by supply chain practitioners, helping them make decisions that can mitigate or prevent risks from occurring. In this work, we first propose a supply chain risk prediction framework using data-driven AI techniques and relying on the synergy between AI and supply chain experts. We then explore the trade-off between prediction performance and interpretability by implementing and applying the framework on the case of predicting delivery delays in a real-world multi-tier manufacturing supply chain. Experiment results show that prioritising interpretability over performance may require a level of compromise, especially with regard to average precision scores.",
keywords = "Supply chain risk management, Risk analysis, Risk prediction, Machine learning, Interpretability",
author = "George Baryannis and Samir Dani and Grigoris Antoniou",
year = "2019",
month = "7",
day = "29",
doi = "10.1016/j.future.2019.07.059",
language = "English",
volume = "101",
pages = "993--1004",
journal = "Future Generation Computer Systems",
issn = "0167-739X",
publisher = "Elsevier",

}

TY - JOUR

T1 - Predicting supply chain risks using machine learning

T2 - The trade-off between performance and interpretability

AU - Baryannis, George

AU - Dani, Samir

AU - Antoniou, Grigoris

PY - 2019/7/29

Y1 - 2019/7/29

N2 - Managing supply chain risks has received increased attention in recent years, aiming to shield supply chains from disruptions by predicting their occurrence and mitigating their adverse effects. At the same time, the resurgence of Artificial Intelligence (AI) has led to the investigation of machine learning techniques and their applicability in supply chain risk management. However, most works focus on prediction performance and neglect the importance of interpretability so that results can be understood by supply chain practitioners, helping them make decisions that can mitigate or prevent risks from occurring. In this work, we first propose a supply chain risk prediction framework using data-driven AI techniques and relying on the synergy between AI and supply chain experts. We then explore the trade-off between prediction performance and interpretability by implementing and applying the framework on the case of predicting delivery delays in a real-world multi-tier manufacturing supply chain. Experiment results show that prioritising interpretability over performance may require a level of compromise, especially with regard to average precision scores.

AB - Managing supply chain risks has received increased attention in recent years, aiming to shield supply chains from disruptions by predicting their occurrence and mitigating their adverse effects. At the same time, the resurgence of Artificial Intelligence (AI) has led to the investigation of machine learning techniques and their applicability in supply chain risk management. However, most works focus on prediction performance and neglect the importance of interpretability so that results can be understood by supply chain practitioners, helping them make decisions that can mitigate or prevent risks from occurring. In this work, we first propose a supply chain risk prediction framework using data-driven AI techniques and relying on the synergy between AI and supply chain experts. We then explore the trade-off between prediction performance and interpretability by implementing and applying the framework on the case of predicting delivery delays in a real-world multi-tier manufacturing supply chain. Experiment results show that prioritising interpretability over performance may require a level of compromise, especially with regard to average precision scores.

KW - Supply chain risk management

KW - Risk analysis

KW - Risk prediction

KW - Machine learning

KW - Interpretability

UR - http://www.scopus.com/inward/record.url?scp=85069864648&partnerID=8YFLogxK

U2 - 10.1016/j.future.2019.07.059

DO - 10.1016/j.future.2019.07.059

M3 - Article

VL - 101

SP - 993

EP - 1004

JO - Future Generation Computer Systems

JF - Future Generation Computer Systems

SN - 0167-739X

ER -