Prioritizing barriers to institutional artificial intelligence adoption using Fuzzy FUCOM and data-driven segmentation
Computers and Industrial Engineering, cilt.221, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 221
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.cie.2026.112301
- Dergi Adı: Computers and Industrial Engineering
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, ABI/INFORM, Aerospace Database, Applied Science & Technology Source, Compendex, INSPEC, DIALNET, Business Source Ultimate (EBSCO), Engineering Source (EBSCO), Technology Collection (ProQuest)
- Anahtar Kelimeler: Artificial Intelligence (AI) adoption, Fuzzy FUCOM, Institutional decision making, K-means clustering, Public administration, Segmentation analysis
- İstanbul Gelişim Üniversitesi Adresli: Evet
Özet
Artificial Intelligence (AI) has strong potential to improve decision-making and service delivery in public administration, yet its institutional adoption remains slower than expected due to multi-dimensional barriers. This study prioritizes and profiles the key barriers to AI adoption in public institutions through a two-stage analytical framework. First, fifteen barriers identified via targeted literature review and expert consultation are weighted using the Fuzzy Full Consistency Method (Fuzzy FUCOM) to obtain a consistent priority structure. Second, perceptions of 21 public-sector professionals and related stakeholders are segmented via K-means clustering based on Likert-scale evaluations, yielding two exploratory barrier-perception profiles. The Fuzzy FUCOM results indicate a strong financial–structural dominance, with insufficient funding, high installation and maintenance costs, and inadequate infrastructure accounting for nearly 45 % of total barrier weight. Mid-ranked barriers emphasize trust and compliance concerns, including cybersecurity and personal data protection, alongside change-related factors such as resistance and algorithmic opacity. The segmentation analysis further demonstrates that barrier salience differs across perception profiles, indicating a difference between the aggregate FUCOM priority structure and segment-specific emphasis patterns. By combining fuzzy MCDM-based prioritization with data-driven segmentation, the study provides an empirically grounded roadmap for public-sector AI transformation and highlights the need for policy designs that jointly address structural readiness and governance legitimacy.