Journal of Information Technology Review Gong Ning 17 1 2026 https://doi.org/10.6025/jitr/2026/17/1/10-21 https://www.dline.info/jitr/fulltext/v17n1/jitrv17n1_2.pdf Customer value evaluation is a critical task for new energy vehicle enterprises seeking to optimize resource allocation, improve customer relationship management, and enhance long term profitability under increasingly competitive market conditions. Traditional customer value models often rely on static parameters and heuristic weighting schemes, which struggle to scale effectively under large customer volumes and high dimensional data. To address these limitations, this paper proposes a customer value evaluation model based on an improved genetic algorithm (GA). Building on the classical GA framework, an adaptive genetic algorithm is introduced in which crossover and mutation probabilities are dynamically adjusted according to population fitness characteristics, thereby enhancing global search capability and convergence stability. The proposed model encodes customer characteristic parameters as chromosomes and employs a fitnessdriven evolutionary process to identify optimal parameter combinations for customer value assessment. Simulation experiments conducted on a representative new energy vehicle enterprise dataset demonstrate that the improved genetic algorithm converges faster and achieves higher accuracy than the traditional GA. The results confirm that the proposed approach effectively improves the precision and robustness of customer value evaluation, providing a scalable and intelligent decision support tool for enterprise customer management.