Ka-means

Decentralized multi-robot missions require task discovery, communication, capability-aware allocation, temporary coalition formation, synchronized execution, and recovery from failed assignments. Incentives are often proposed as a mechanism for aligning local decisions with mission outcomes, but their effectiveness depends on whether the system can first establish informational and operational feasibility. This paper introduces MOSAIC (Mission-Oriented Self-Organization through Auctions, Incentives, and Coalitions), an agent-based model of heterogeneous robots operating under partial observability, spatially variable risk, limited energy, dynamic communication, and individual and cooperative task requirements. Seven paired-seed experiments comprising 690 official runs evaluate baseline viability, reward regimes, communication structure, capability heterogeneity, cooperative-task demand, reputation and adaptive strategies, and mission-incentive strength. The strongest determinants of completion are structural: high cooperative demand reduces completion by 23.22 percentage points, moderate capability heterogeneity outperforms both homogeneous and highly heterogeneous teams by 5.94 percentage points, and increasing communication radius from 4 to 8 adds 3.56 percentage points while improving information coverage, coalition delay, coordination overhead, and mission efficiency. By contrast, leader-priority rewards increase inequality and auction congestion without improving completion, and stronger mission incentives increase payouts without producing reliable mission-performance gains. The evidence indicates that incentives cannot compensate for missing information, fragmented capability matching, or insufficient coalition-formation capacity. This manuscript has not yet undergone journal peer review.

Customer churn in voluntary health insurance is a multidimensional decision problem that resists reduction to a single probability score. This paper proposes an integrated four-layer analytical framework combining topological behavioral segmentation (UMAP + HDBSCAN), explainable churn prediction (XGBoost + SHAP), survival analysis (Kaplan-Meier, Cox proportional hazards, Random Survival Forest), and financial simulation. The framework is validated on administrative records from a Colombian health insurer covering approximately 298,000 complementary-plan subscribers (1,019,505 subscription-period records after censoring filters; 40 predictors; stratified 80/20 split). XGBoost achieves AUC = 0.86 (KS = 56.3%, Lift@10% = 3.29, Brier = 0.125), compared with 0.80 for logistic regression; the Random Survival Forest yields C-index = 0.818. Topological segmentation uncovers two structural phenomena: a Health Paradox, in which young, healthy, low-utilization subscribers exhibit the highest cancellation volatility, and a Fortress Effect, in which utilization of the insurer’s own provider network is associated with a 31.6% lower instantaneous cancellation hazard (Cox HR = 0.684, p < 0.001). Financial simulation under baseline assumptions projects a net annual benefit of COP $5,732 million (ROI 302%). The central contribution lies in the integration of analytical layers that jointly address identity, causation, timing, and atypical cases in churn dynamics.

MOSAIC is an agent-based NetLogo model of decentralized mission coordination among heterogeneous robots operating under partial observability, limited energy, spatially variable risk, dynamic communication, and individual and cooperative task requirements. Robots discover tasks locally, exchange task information through temporary communication links, submit capability-, energy-, deadline-, and risk-aware bids, compete for individual contracts, and form temporary coalitions for cooperative tasks. The model integrates decentralized auctions, greedy capability-based coalition formation, contract release and reassignment, four reward regimes, reputation, adaptive bidding strategies, failure traceability, and mission-, network-, information-, inequality-, and coalition-level metrics, and operates without a centralized mission planner or global combinatorial assignment solver. Seven paired-seed BehaviorSpace experiments comprising 690 official simulation runs evaluate baseline mission viability, reward regimes, communication structure, capability heterogeneity, cooperative-task demand, reputation and adaptive strategies, and mission-incentive strength. The results indicate that structural coordination capacity, particularly information reach, capability compatibility, and feasible coalition construction, has a stronger effect on mission completion than increasing incentive intensity within the tested architecture and parameter ranges. MOSAIC includes nine automated verification invariants covering task-state consistency, contract consistency, energy accounting, reward accounting, coalition membership, failure traceability, knowledge integrity, reputation bounds, and strategy validity.