2024
Hidayat, Rahmat; Ourairat, Apivut; Wicaksono, Hendro
Explainable Artificial Intelligence in Agrifood Supply Chain: State of the Art Review Proceedings Article
In: Lecture Notes in Mechanical Engineering : Flexible Automation and Intelligent Manufacturing: Manufacturing Innovation and Preparedness for the Changing World Order, Springer, 2024.
Abstract | Links | BibTeX | Tags: agrifood, artificial intelligence, explainable AI, supply chain management
@inproceedings{nokey,
title = {Explainable Artificial Intelligence in Agrifood Supply Chain: State of the Art Review},
author = {Rahmat Hidayat and Apivut Ourairat and Hendro Wicaksono
},
doi = {https://doi.org/10.1007/978-3-031-74485-3_33},
year = {2024},
date = {2024-12-13},
booktitle = {Lecture Notes in Mechanical Engineering : Flexible Automation and Intelligent Manufacturing: Manufacturing Innovation and Preparedness for the Changing World Order},
publisher = {Springer},
abstract = {The increasing pressure to feed a growing population of humans for food security, with constantly changing food consumption behavior, as well as in recent light of livestock treatment and awareness for food sustainability both economically and ecologically, also due to the challenges of climate change, lead to challenges for the industries operating in the Agrifood Supply Chain (ASC). Recent technological strides in Data Analysis, Internet of Things (IoT), Machine Learning (ML), and Artificial Intelligence (AI) have ushered in a digitized and intelligent era within the ASC, reshaping production quality, sustainability, and food longevity. However, the nascent stage of AI and ML methods within the ASC raises questions about their reliability, value, transparency, and understandability. The prevalent use of black box methods underscores the need for more explainable methodologies, as the opacity of current approaches restricts widespread applicability. This paper presents a State-of-the-Art Review of Explainable AI (XAI) and ML methods in the ASC, delving into operations spanning “Farm-to-Fork,” encompassing agriculture production, processes, quality assurance, tracking, warehousing, distribution, packaging, retailing, safety, and sustainability. The research identifies challenges and proposes research directions, offering researchers an overview of opportunities to implement XAI methods in the ASC. The exploration of coexisting problems and their solutions enhances our understanding of intelligent systems in the ASC, providing valuable insights for stakeholders’ decision-making processes.
},
keywords = {agrifood, artificial intelligence, explainable AI, supply chain management},
pubstate = {published},
tppubtype = {inproceedings}
}
The increasing pressure to feed a growing population of humans for food security, with constantly changing food consumption behavior, as well as in recent light of livestock treatment and awareness for food sustainability both economically and ecologically, also due to the challenges of climate change, lead to challenges for the industries operating in the Agrifood Supply Chain (ASC). Recent technological strides in Data Analysis, Internet of Things (IoT), Machine Learning (ML), and Artificial Intelligence (AI) have ushered in a digitized and intelligent era within the ASC, reshaping production quality, sustainability, and food longevity. However, the nascent stage of AI and ML methods within the ASC raises questions about their reliability, value, transparency, and understandability. The prevalent use of black box methods underscores the need for more explainable methodologies, as the opacity of current approaches restricts widespread applicability. This paper presents a State-of-the-Art Review of Explainable AI (XAI) and ML methods in the ASC, delving into operations spanning “Farm-to-Fork,” encompassing agriculture production, processes, quality assurance, tracking, warehousing, distribution, packaging, retailing, safety, and sustainability. The research identifies challenges and proposes research directions, offering researchers an overview of opportunities to implement XAI methods in the ASC. The exploration of coexisting problems and their solutions enhances our understanding of intelligent systems in the ASC, providing valuable insights for stakeholders’ decision-making processes.