Europe is experiencing a critical driver shortage due to demanding working conditions and high turnover rates. The MILESTONE project addresses this issue by leveraging mobile and technological solutions to improve the health and working conditions of occupational drivers. This study utilized commercial wearables, specifically the Fitbit Charge 5 and Zephyr Bioharness 3.0, to detect acute mental stress in 33 occupational drivers. Participants underwent stress-inducing tasks under various conditions, with stress levels measured using the Visual Analog Scale (VAS). Personalized machine learning models, particularly random forest (RF) classifiers, were developed to predict stress from wearable data inputs. Results showed that Zephyr-based models were about 22% more accurate than Fitbit-based models, with further accuracy gains when using multimodal data inputs. Significant correlations were identified between subjective stress levels and factors such as BMI, sleep quality, and sleepiness. Despite the limitation of limited data per participant, the study demonstrated the feasibility of using affordable wearables for stress detection. These findings suggest potential interventions to enhance drivers’ working conditions and health, emphasizing the need for future research to validate these results on a larger scale and to explore advanced wearable technologies for comprehensive stress monitoring. Lastly, this study highlights the importance of addressing sleep quality and other personal factors in managing stress and improving driver safety.