Optimization of Green Consumer Awareness Communication Budget Allocation Using Genetic Algorithms: A Partially Calibrated Model Based on Published Secondary Data

Abstract

This study proposes a Genetic Algorithm (GA)-based framework for optimizing six marketing-mix variables: price, delivery experience, brand trust, sustainable packaging, digital UI/UX experience, and ESG budget. The model integrates published secondary statistics to calibrate key parameters while employing illustrative assumptions for variables lacking publicly available data. Results show that the GA converges to a stable interior solution, improving fitness by 14.1% over the initial generation. Sensitivity analysis indicates that the optimal allocation remains robust across a wide range of utility-cost trade-offs, although lower weight on utility leads to corner solutions. A comparative evaluation with Particle Swarm Optimization (PSO) and Differential Evolution (DE) reveals that DE achieves comparable solution quality with faster convergence, whereas PSO converges more rapidly but exhibits greater variability. The study demonstrates the feasibility of combining secondary market evidence with evolutionary optimization and provides a reproducible framework for marketing resource-allocation decisions. It also highlights methodological limitations associated with aggregating heterogeneous survey statistics into a unified weighting scheme, offering directions for future empirical validation.