PURPOSES : Since the 1990s, underground utility projects have been conducted to solve the problem of aerial communication cables. The purpose of this study is to derive optimal measures for preventing collisions with existing underground utilities and for future maintenance in the implementation of the utilities undergrounding projects. This study considered the identifier sensor and tested the optimal sensor performance for more accurate and systematic management. METHODS : In this study, three representative technologies were selected from identifier sensors generally used in air and the possibility for their use in soil and asphalt was confirmed by simulating the environment via a test construction. Three identifier sensors were selected: BLE (Bluetooth low energy) beacon, ultra-high frequency radio frequency identification (UHF RFID), and a geomagnetic recognizer. The long-term recognition performance of each identifier sensor was tested using the underground depth as a variable and the results were analyzed for comparison. RESULTS : The results of the test under limited conditions and environment demonstrated that the BLE Beacon had advantages in equipment composition, recognition range, and speed but exhibited problems with batteries in winter. The geomagnetic recognizer did not show the exact location and its influence on the surrounding environment was a disadvantage. CONCLUSIONS : Although the performance of UHF RFID has been demonstrated to be relatively suitable under these test conditions, it seems that the impact of the more diverse installation depth or medium should be reviewed for actual commercialization.

PURPOSES : Despite the availability of larger traffic data and more advanced data collection methods, the problem of missing data is yet to be solved. Imputing missing data to ensure data quality and reliability of statistics has always been challenging. Missing data are imputed via several existing methods, such as autoregressive integrated moving average, exponential smoothing, and interpolation. However, these methods are complicated and results in significant errors. METHODS : A deep-learning method was applied in this study to impute traffic volume data of the South Korean national highway. Traffic data were trained using the long short-term memory method, which is a suitable deep-learning method for time series analysis. RESULTS : Three cases were proposed to estimate the traffic volume. In the first case, which represented the general conditions, the mean absolute percentage error (MAPE) was 12.7%. The second estimation case, which was based on the opposite traffic flow, exhibited a MAPE of 17%~18%. The third case, which was estimated using adjacent-section data, had a MAPE of 18.2%. CONCLUSIONS : Deep learning may be a suitable alternative data imputation method based on the limited site and data. However, its application depends on the specific situation. Furthermore, deep-learning models can be improved using an ensemble method, batch-size, or through model-structure optimization.