GA-MPG: efficient genetic algorithm for improvised mobile plan generation

• Published in: Journal of ambient intelligence and humanized computing Vol. 15; no. 10; pp. 3675 - 3691
• Main Authors: Shukla, Rohan S., Ghuse, Ekta A., Diwan, Tausif, Tembhurne, Jitendra V., Sahare, Parul
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-10-2024; Springer Nature B.V
• Subjects: Artificial Intelligence; Business metrics; Computational Intelligence; Corporate profits; COVID-19; Customer services; Customers; Datasets; Design; Engineering; Genetic algorithms; Optimization; Order quantity; Original Research; Profits; Robotics and Automation; Telecommunications; Telecommunications industry; User Interfaces and Human Computer Interaction; Customer churn; Communication service providers (CSPs); Genetic algorithm; Mobile plans
• ISSN: 1868-5137; 1868-5145
• DOI: 10.1007/s12652-024-04846-3

In the competitive landscape of the telecom sector, a Communication Service Provider's success hinges on its ability to offer compelling mobile plans tailored to diverse customer needs. This not only boosts company profits but also enhances metrics like average revenue per user (ARPU), customer lifecycle value, and reduces customer churn. Striking a balance between these objectives presents a formidable task. To address this challenge, we propose a novel approach called Genetic Algorithm Mobile Plan Generation (GA-MPG). The proposed method stands out for its deterministic approach that equally focuses on minimizing customer churn. This is done by providing them with the best-suited plans without making them pay extra for features they would use. The efficient mobile plan generation using GA-MPG is accomplished by the combination of the AdaBoost classifier and the Fuzzy model. The AdaBoost is utilized for feasible mobile plan generation and predicting the optimal solution amongst the various plans. Additionally, a fuzzy model recommends personalized plans based on customers' typical service usage. This also maximizes company profits, contrasting with existing strategies employed by various telecom companies which focus on one of the two problems. The proposed GA-MPG algorithm demonstrated promising results on a prominent US-based telecom dataset encompassing around 7000 customers, with a substantial 44% reduction in customer churn. These findings are based on the simulation results. The algorithm also shows improvements of 13% and 18% in ARPU and company profit, respectively, over a defined period.