Artificial Intelligence Constrained Multi-Objective Optimisation of Community Health Worker Vaccine Carrier Allocation Under Variable Cold Box Holdover Failure Reporting: A computed Healthcare Immunisation Coverage Equity Evaluation

Vaccine carrier allocation for community health workers (CHWs) in low-resource immunisation programmes must balance coverage, equity and cold-chain reliability, yet cold box holdover failure is rarely reported consistently, and allocation rules that ignore reporting heterogeneity can systematically disadvantage remote populations. the present analysis formulates the allocation problem as a constrained multi-objective optimisation over a computed district of CHW catchments, where each carrier has a holdover failure probability drawn from a reporting-dependent distribution, and where the decision variables assign carriers to catchments under budget and cold-chain constraints. The optimisation is solved with an artificial-intelligence-guided evolutionary algorithm that maintains a Pareto archive over coverage and equity objectives, and the resulting allocations are compared with proportional and needs-based heuristics across 600 replications of N = 500 computed catchment-days. The analysis characterises estimator behaviour under the declared data-generating process (intercept 12.0, exposure 0.42, group -3.1, severity 1.35, error SD 4.2) and reports the computed bias, root-mean-square error and interval coverage for the intercept and exposure coefficients. The framework proposes a reporting-aware allocation rule that would permit equity gains under variable holdover failure reporting, and the paper derives the conditions under which such gains are identifiable from routine data.