International Journal for Applied Information Management

Graph-Driven Knowledge Reasoning for Human AI Collaborative Strategic Decision Support in Supply Chain Networks

2026-08-19T22:11:24+07:00

Strategic decision support in supply chain networks remains constrained by fragmented data structures, limited visibility of multi-tier dependencies, and weak integration between relational analytics and managerial prioritization. This study proposes a graph-driven knowledge reasoning framework that combines knowledge graph construction, graph-based inference, multi-criteria recommendation ranking, and scenario-based strategic evaluation to support decision-making in complex supply chain environments. The modeled network comprised 248 nodes and 612 directed edges representing suppliers, factories, warehouses, transport routes, retailers, and disruption events. Structural analysis showed an average degree of 4.94, with suppliers forming the largest entity class at 72 nodes, while factories exhibited the highest relational intensity with an average degree of 7.4. The reasoning engine identified several high-risk propagation corridors, including the path S12 → F3 → R9 with a risk score of 0.91, indicating concentrated supplier dependence and high downstream delay exposure. Additional critical paths, such as S21 → F6 → W2 and S4 → W7 → T5, recorded scores of 0.84 and 0.81, revealing capacity bottlenecks, bridge-node fragility, and regionally amplified service disruption. Entity-level inference further showed that Warehouse W7 and Route R9 were strategically important despite not being dominant in raw node volume, confirming that graph-based reasoning can reveal hidden structural vulnerabilities overlooked by conventional analysis. In the decision support phase, supplier diversification obtained the highest composite recommendation score at 8.14, followed by inventory repositioning at 7.96 and route redundancy at 7.88. Comparative scoring indicated that supplier diversification produced the strongest resilience value at 8.9, while inventory repositioning achieved the best balance between feasibility and cost efficiency. Scenario analysis demonstrated that diversification performance increased from 7.6 under normal conditions to 8.8 under severe disruption, whereas inventory repositioning declined from 8.0 to 7.1, indicating that short-term stabilization strategies lose effectiveness under deeper structural stress. Expert-oriented validation also reported high managerial acceptance, with average scores of 4.6 for strategic relevance, 4.5 for interpretability, and 4.4 for practical usefulness. These findings show that graph-driven knowledge reasoning can improve strategic visibility, detect hidden dependencies, and generate interpretable, context-sensitive recommendations for supply chain decision support.

Dynamic Knowledge Intelligence for Human AI Collaborative Operational Decision Making Across Distributed Enterprise Data Environments

2026-08-20T17:59:52+07:00

Distributed enterprise environments frequently contain fragmented operational data distributed across ERP platforms, CRM systems, workflow logs, IoT streams, and unstructured document repositories, which reduces the speed and consistency of managerial response. This study proposes a dynamic knowledge intelligence framework that integrates heterogeneous enterprise data, transforms them into a temporal contextual knowledge layer, and connects that layer to an operational decision engine for action prioritization. The framework was evaluated across five analytical dimensions, namely integration quality, knowledge graph enrichment, recommendation performance, response efficiency, and cross-domain business impact. The harmonization stage improved integration consistency from 0.61 to 0.89 for ERP data, from 0.58 to 0.85 for CRM data, from 0.64 to 0.91 for workflow logs, from 0.47 to 0.78 for documents, and from 0.55 to 0.84 for IoT streams, with error reduction ranging from 26% to 34%. In the contextual intelligence layer, knowledge graph nodes increased from 120 at T1 to 352 at T5, while relations expanded from 210 to 812, indicating that contextual linkage grew faster than object accumulation. The decision engine produced strong recommendation performance across five enterprise scenarios, with accuracy values of 0.88 for backlog control, 0.91 for inventory risk, 0.86 for escalation routing, 0.83 for policy conflict, and 0.89 for resource allocation, while expert alignment ranged from 0.81 to 0.90. Efficiency analysis showed average decision latency of 0.82 seconds at 50 concurrent requests, 1.04 seconds at 100 requests, 1.36 seconds at 200 requests, 1.95 seconds at 400 requests, and 2.88 seconds at 800 requests, confirming stable operational responsiveness under low to heavy workloads and controlled degradation only under stress-level concurrency. Business impact analysis further showed differentiated strategic gains, including 28% higher response speed in service management, 31% stronger risk mitigation in supply chain settings, 33% better decision traceability in governance contexts, 20% stronger coordination quality in resource alignment, and 24% better decision consistency in enterprise-wide operations. These findings demonstrate that the proposed framework effectively converts fragmented enterprise data into contextual, updateable, and decision-ready intelligence, thereby improving operational agility, recommendation credibility, and cross-domain decision value in distributed enterprise environments.

Explainable Human AI Decision Intelligence for Executive Strategy Development Using Enterprise Documents and Real-Time Business Signals

2026-08-20T18:10:05+07:00

The growing complexity of enterprise environments has increased the need for decision intelligence systems that can integrate organizational knowledge, real-time business conditions, and transparent reasoning into executive strategy development. This study proposes an explainable decision intelligence framework that combines enterprise documents, live business signals, contextual retrieval, evidence fusion, strategic prioritization, and traceable explanation generation within a unified architecture. The framework was evaluated through five analytical dimensions covering retrieval quality, contextual fusion, recommendation strength, explainability performance, and executive validation. The results showed that signal-aware retrieval consistently outperformed document-only retrieval across six executive query categories, with Precision@5 increasing from 0.67-0.74 to 0.80-0.86. Growth slowdown achieved the highest retrieval performance at 0.86, followed by operational delay at 0.85 and cost pressure at 0.84. Contextual fusion analysis further indicated strong document-signal alignment in finance and risk domains, with context link strengths of 0.88 and 0.85, respectively, while bundle density reached 14 evidence items in finance and 13 in risk. In the recommendation layer, operational stabilization received the highest strategic priority score at 0.87, followed by cost restructuring at 0.82 and selective expansion at 0.78, showing that the framework could differentiate between corrective, defensive, and growth-oriented strategies according to current business conditions. Explainability results revealed that recommendation logic was distributed across multiple factors, with margin pressure contributing 29% of explanation weight, execution delay 24%, demand instability 18%, risk exposure 15%, and strategic fit 14%, indicating balanced and evidence-grounded reasoning rather than single-factor dominance. Executive validation also produced strong outcomes, with average ratings of 4.8 for decision support, 4.7 for evidence grounding, 4.6 for clarity, 4.5 for trust, and 4.4 for actionability on a five-point scale. These findings demonstrate that the proposed framework improves strategic decision support by linking enterprise memory with live business signals and by presenting prioritized recommendations with transparent evidence trails. The study contributes a practical model for executive strategy development in dynamic enterprise settings, while also extending the literature on explainable AI, enterprise retrieval-augmented intelligence, and decision support system design.

A Human AI Collaborative Knowledge Mining Framework for Enterprise Risk Detection and Managerial Decision Intelligence

2026-08-20T18:36:51+07:00

Enterprise risk detection increasingly requires analytical frameworks capable of integrating structured records, semi-structured logs, and unstructured organizational text into a unified managerial intelligence process. This study proposes a hybrid knowledge mining framework designed to detect enterprise risk patterns and transform them into prioritized, explanation-ready decision support for managers. The framework combines data preprocessing, semantic knowledge extraction, predictive risk scoring, hybrid evidence fusion, and managerial prioritization within one architecture. Experimental results showed that the proposed framework achieved an accuracy of 0.91, precision of 0.89, recall of 0.87, and F1-score of 0.88, outperforming the structured-only model with 0.82 accuracy and 0.78 F1-score, the text-only model with 0.79 accuracy and 0.75 F1-score, and the log-only model with 0.81 accuracy and 0.78 F1-score. Incremental fusion analysis further demonstrated that the full hybrid configuration produced the strongest performance, exceeding structured-only by 0.13 points in F1-score and surpassing pairwise combinations such as structured plus logs at 0.85 and structured plus text at 0.86. Class-level evaluation showed that the framework performed best on financial irregularity with precision 0.92, recall 0.89, and F1-score 0.90, followed by compliance deviation with F1-score 0.89 and operational delay with F1-score 0.86, while still maintaining reliable performance for supplier instability at 0.82 and communication risk at 0.81. The managerial prioritization layer generated differentiated action recommendations, distributing 44.3% of cases to monitoring, 29.5% to mitigation, 16.2% to escalation, and 10.0% to review, indicating that the framework can triage organizational risks without inflating urgency. Expert-oriented evaluation of explanation quality also showed strong operational usefulness, with average scores of 4.5 for clarity, 4.4 for traceability, 4.6 for relevance, 4.3 for actionability, and 4.4 for trust on a five-point scale. These findings indicate that the proposed framework extends enterprise risk analytics beyond standalone prediction by combining heterogeneous evidence fusion, contextual reasoning, and managerial decision intelligence in a single system. The study contributes a practical and scalable architecture for organizations seeking to convert fragmented enterprise data into actionable and interpretable risk governance.

Human AI Collaborative Knowledge Synthesis for Corporate Policy Analysis and Cross-Division Decision Alignment

2026-08-20T18:45:44+07:00

Corporate policy environments are often characterized by fragmented documents, inconsistent terminology, overlapping authority, and division-specific procedural interpretations that weaken organizational alignment. This study proposes an intelligent knowledge synthesis framework to analyze corporate policy corpora and assess cross-division decision alignment through an interpretable semantic pipeline. The framework integrates document acquisition, preprocessing, policy representation, knowledge synthesis, contradiction detection, alignment scoring, and evaluation within a unified architecture. Using a corpus of 540 source documents consisting of corporate policies, standard operating procedures, internal memoranda, strategic directives, and implementation notes, the method transformed the data into 1,980 normalized policy units and generated 246 synthesized policy clusters with a mean semantic coherence of 0.87. Domain-level results showed the highest coherence in compliance (0.90), approval governance (0.89), and reporting (0.88), while resource allocation and risk control produced lower but still stable coherence scores of 0.84 and 0.86. Contradiction analysis identified 173 significant conflict cases and 94 structured exception relationships, with conditional exceptions representing the largest pattern, followed by direct conflict, authority clash, and procedural override. Inter-divisional analysis showed the strongest friction between operations and compliance, IT and legal, and finance and operations, indicating that policy misalignment frequently emerges from localized procedural adaptation rather than explicit policy rejection. Across 320 evaluated decision cases, the alignment model achieved a mean score of 0.84, with 71.6% of cases classified in the high-alignment band, 19.1% in the moderate band, 6.9% in the low band, and 2.5% in the critical band. Category-level evaluation showed the highest alignment for compliance (0.91) and reporting (0.89), while resource-related decisions recorded the lowest score at 0.77. Division-level analysis further revealed that legal and compliance maintained the strongest stability profiles, with mean alignment scores of 0.90 and 0.88 and low variance values of 0.012 and 0.015, whereas IT and operations displayed greater volatility, with mean scores of 0.80 and 0.78 and variance values of 0.031 and 0.036. Comparative evaluation demonstrated that the full framework outperformed the baseline and synthesis-only configurations across synthesis quality, conflict detection, alignment stability, and interpretive usefulness, reaching scores of 0.90, 0.87, 0.89, and 0.92, respectively. These findings show that intelligent knowledge synthesis can function as a robust analytical foundation for enterprise policy interpretation, contradiction diagnosis, and coordination-aware decision support across organizational divisions.