Artificial Intelligence and Innovation to Reduce the Impact of Extreme Weather Events on Sustainable Production

• Main Authors: Effah, Derrick, Bai, Chunguang, Quayson, Matthew
• Format: Journal Article
• Language: English
• Published: 21-09-2022
• Subjects: Computer Science - Artificial Intelligence; Computer Science - Computers and Society
• DOI: 10.48550/arxiv.2210.08962

Frequent occurrences of extreme weather events substantially impact the lives
of the less privileged in our societies, particularly in agriculture-inclined
economies. The unpredictability of extreme fires, floods, drought, cyclones,
and others endangers sustainable production and life on land (SDG goal 15),
which translates into food insecurity and poorer populations. Fortunately,
modern technologies such as Artificial Intelligent (AI), the Internet of Things
(IoT), blockchain, 3D printing, and virtual and augmented reality (VR and AR)
are promising to reduce the risk and impact of extreme weather in our
societies. However, research directions on how these technologies could help
reduce the impact of extreme weather are unclear. This makes it challenging to
emploring digital technologies within the spheres of extreme weather. In this
paper, we employed the Delphi Best Worst method and Machine learning approaches
to identify and assess the push factors of technology. The BWM evaluation
revealed that predictive nature was AI's most important criterion and role,
while the mass-market potential was the less important criterion. Based on this
outcome, we tested the predictive ability of machine elarning on a publilcly
available dataset to affrm the predictive rols of AI. We presented the
managerial and methodological implications of the study, which are crucial for
research and practice. The methodology utilized in this study could aid
decision-makers in devising strategies and interventions to safeguard
sustainable production. This will also facilitate allocating scarce resources
and investment in improving AI techniques to reduce the adverse impacts of
extreme events. Correspondingly, we put forward the limitations of this, which
necessitate future research.