The quick expansion of the transportation sector has intensified the competitiveness among cab drivers, spurring the implementation of creative tactics to enhance profitability. In this research, we propose a strategy employing deep reinforcement learning (DRL) to develop a recommendation system tailored for cab drivers. Unlike conventional recommendation systems, which usually depend on static rules or predefined algorithms, our DRL-based system dynamically learns optimum suggestions via interactions with the environment, therefore adapting to the evolving situations of both drivers and passengers. By characterizing the recommendation problem as a reinforcement learning issue, we create an agent capable of making knowledgeable judgments on route and passenger selection to maximize the driver’s profit while ensuring customer pleasure. We assess the effectiveness of our technique, indicating its utility in enhancing driver earnings compared to standard methods. The findings highlight the viability of mixing contemporary machine learning algorithms with realistic transportation systems to handle challenging optimization issues and enable cab drivers to thrive in today’s competitive market.