Trusted AI Decision Support for Military Logistics and Sustainment in Contested Environments - Softcover

Pokorny, Dr. Laszlo

Softcover

ISBN 13: 9798258915269

Verlag: Independently published, 2026

Inhaltsangabe

This study investigated how explainable artificial intelligence (XAI) modalities affect trust calibration in AI-enabled military logistics decision support across operational contexts and cultural backgrounds. A four-phase sequential mixed-methods design was employed. Phase 1 used cognitive task analysis to identify 10 critical decision nodes spanning tactical, operational, and strategic military logistics domains, each characterized by distinct information requirements, time pressures, and XAI modality needs. Phase 2 developed an AI prototype integrating World Bank Logistics Performance Index data (N = 1,407 records across 217 countries), USAspending defense procurement records ($127.08 billion across 5,808 contracts), and Armed Conflict Location and Event Data Project conflict data (36,248 country-year observations), achieving 83.3% classification accuracy for logistics risk prediction with three XAI output modalities: SHAP feature importance, counterfactual explanations, and confidence intervals. Phase 3 conducted a 3 × 3 between-subjects factorial experiment (N = 135) examining XAI modality and operational context (permissive, degraded, contested) effects on decision quality, trust calibration, explanation satisfaction, and situational awareness among multinational military logistics professionals from France, India, Japan, and the United States. Results revealed that operational context was the primary determinant of trust calibration and situational awareness, while XAI modality was the primary determinant of explanation satisfaction. A significant XAI × context interaction on trust calibration revealed that SHAP explanations excelled in contested environments while counterfactual explanations excelled in permissive settings. Cross-cultural moderation was significant (p = .045). Phase 4 employed a 12-week longitudinal field study (N = 45) using hierarchical linear modeling, revealing that trust followed a three-phase trajectory (Calibration, Stabilization, Mature Trust) with counterfactual XAI yielding the highest growth rate (b = 0.169/week). The findings demonstrate that context-adaptive XAI is essential for military logistics AI, no single explanation modality is universally optimal, and culturally configurable interfaces are necessary for effective multinational operations.

Details zur Ausgabe

• Verlag: Independently published
• Erscheinungsdatum: 2026
• Sprache: Englisch
• ISBN 13: 9798258915269
• Einband: Taschenbuch
• Anzahl der Seiten: 326
• Kontakt zum Hersteller: Manufactured by Amazon on behalf of the author; https://www.amazon.de/hz/contact-us; ; c/o Amazon Media EU S.à.r.l., 38 Avenue John F. Kennedy; Luxembourg; L-1855; Luxemburg