The Impact of Artificial Intelligence Applications on Improving Electronic Administrative Decision-Making: A Field Study in Public Hospitals Operating in Benghazi

Abstract

This study aimed to investigate the impact of implementing artificial intelligence (AI) in its three dimensions (data analysis, administrative automation, and decision support) on improving the electronic administrative decision-making process in public hospitals in Benghazi. The study employed a descriptive-analytical approach and used a questionnaire to collect data from 95 managers in senior and middle management positions across Benghazi's public hospitals using a census. Data were analyzed using SPSS software. The results showed that the overall level of AI implementation in Benghazi's public hospitals was low (mean ≈ 2.26–2.30, relative weight ≈ 45.2%), with low to very low levels across all dimensions (data analysis: ≈ 45.2%, administrative automation: ≈ ​​44.8%, decision support: ≈ 46.0%). Similarly, the electronic administrative decision-making process was found to be at a low level (arithmetic mean ≈ 2.22, relative weight ≈ 44.4%), with a clear weakness in reliance on accurate numerical data, evaluation of alternatives, acceleration of procedures, reduction of errors, or reduction of personal bias. Statistical results (regression test) confirmed the absence of a statistically significant effect at the (α ≤ 0.05) level of applying artificial intelligence (as a whole or its three dimensions) on the electronic administrative decision-making process. The correlation coefficient (R) ranged between 0.233 and 0.312 (very weak), the coefficient of determination (R²) between 0.054 and 0.097 (only 5.4%–9.7% of the variances explained by artificial intelligence), and the significance values ​​(Sig) between 0.142 and 0.231 (all > 0.05), leading to the acceptance of all null hypotheses. These findings indicate that public hospitals in Benghazi are still in the very early stages of digital transformation, and that artificial intelligence (AI) technologies are not effectively integrated into administrative processes, with a continued heavy reliance on traditional methods and human expertise. The study also reveals a significant gap between the theoretical potential of AI to improve healthcare management and the practical reality in these institutions..