This research aims to analyze the use of SIMON TKBM, identify obstacles in its implementation, and evaluate its impact on the efficiency of loading and unloading activities at Tanjung Perak Port. A quantitative approach was used involving a population consisting of 1480 loading and unloading workers (TKBM) at the port. The research sample was calculated using the Slovin formula with a margin of error of 10%, resulting in a total of 90 representative respondents. Data was collected through a questionnaire with 17 statements and then analyzed using validity, reliability and multiple linear regression tests to identify the relationship between the use of SIMON TKBM, implementation obstacles and its impact on loading and unloading efficiency. Referring to the validity and reliability test results, it shows that all questionnaire items are valid and reliable (Cronbach's Alpha = 0.893). Multiple linear regression analysis shows that the System (X1) and User (X2) variables have a positive effect on Interaction (Y), with the largest contribution coming from the user factor (coefficient = 0.301; significance = 0.003). Meanwhile, the System variable has a low and insignificant influence (coefficient = 0.090; significance = 0.395), which indicates the existence of technical problems in the system. The multiple correlation coefficient value of 0.519 indicates a moderate relationship between the three variables. These findings conclude that the implementation of SIMON TKBM is quite effective in supporting loading and unloading activities, especially with active user support, although technical improvements and identification of other external factors that influence system effectiveness are still needed.

Security systems in the maritime sector are crucial for preventing criminal activities and theft on board ships. The physical key systems commonly used today have vulnerabilities that can be exploited by unauthorized individuals. This research aims to design and test a smart door lock system using facial recognition with the ESP32 Wrover cam and a 4x4 Keypad as an additional security measure. The test results show that the system successfully recognizes faces with a 90% success rate at a distance of 15-100 cm, although there is a decrease in efficiency when more than one face is registered. The system also demonstrated 100% success in