Recently, artificial intelligence (AI) technologies have been employed to predict construction and demolition (C&D) waste generation. However, most studies have used machine learning models with continuous data input variables, applying algorithms, such as artificial neural networks, adaptive neuro-fuzzy inference systems, support vector machines, linear regression analysis, decision trees, and genetic algorithms. Therefore, machine learning algorithms may not perform as well when applied to categorical data. This article uses machine learning algorithms to predict C&D waste generation from a dataset, as a way to improve the accuracy of waste management in C&D facilities. These datasets include categorical (e.g., region, building structure, building use, wall material, and roofing material), and continuous data (particularly, gloss floor area), and a random forest (RF) algorithm was used. Results indicate that RF is an adequate machine learning algorithm for a small dataset consisting of categorical data, and even with a small dataset, an adequate prediction model can be developed. Despite the small dataset, the predictive performance according to the demolition waste (DW) type was R (Pearson’s correlation coefficient) = 0.691–0.871, R² (coefficient of determination) = 0.554–0.800, showing stable prediction performance. High prediction performance was observed using three (for mortar), five (for other DW types), or six (for concrete) input variables. This study is significant because the proposed RF model can predict DW generation using a small amount of data. Additionally, it demonstrates the possibility of applying AI to multi-purpose DW management.

This study examined a method to reduce energy consumption in office buildings. Correspondingly, an optimal control method was proposed for heating, ventilation, and air conditioning (HVAC) systems via two control algorithms that considered the indoor thermal environment. The control algorithms were developed by considering temperature and humidity as the factors of the indoor thermal environment that influence the control of HVAC systems and the predicted mean vote comfort ranges. Furthermore, an experiment was performed using office equipment that incorporated the two control algorithms for HVAC systems, and the correlation between changes in the thermal environment within the office and the occupant’s comfort levels was estimated via an actual survey. The results demonstrated that the proposed control method for HVAC systems, which considered the comfort ranges of temperature and humidity and the thermal adaptation capability, can efficiently maintain the occupant’s comfort with lower energy usage compared with conventional HVAC systems. Thus, the use of the control method contributes to the reduction of total energy consumption in buildings with HVAC systems.

Recently, building automation system (BAS) and building energy management system (BEMS) technologies have been applied to efficiently reduce the energy consumption of buildings. In addition, studies on utilizing large quantities of building data have been actively conducted using artificial intelligence and machine learning. However, the high cost and installation difficulties limit the use of measuring devices to sense the indoor environment of all buildings. Therefore, this study developed a comprehensive indoor environment sensor module with relatively inexpensive sensors to measure the indoor environment of a university building. In addition, an algorithm for predicting the load in real time through machine learning based on indoor environment measurement is proposed. When the reliability of the algorithm for predicting the number of occupants and load according to the indoor CO₂ concentration was quantitatively assessed, the mean squared error (MSE), root mean square deviation (RMSD), and mean absolute error (MAE) were calculated to be 23.1, 4.8, and 2.5, respectively, indicating the high accuracy of the algorithm. Since the sensor used in this study is economical and can be easily applied to existing buildings, it is expected to be favorable for the dissemination of load prediction technology.

Recently, there has been increasing attention on the use of renewable energy in buildings, particularly, the photovoltaic thermal (PVT) system that uses both solar power and thermal energy. However, there is a limit to adopting the PVT system in real buildings because many architects value the aesthetics of buildings or spaces. This study developed a curtain-wall-type liquid-type PVT (CW-PVT) that can be installed on a wall as it integrates with the building. To analyze the system performance, a real-scale experimental plant was established in an outdoor environment. The performance of the CW-PVT unit was verified for two different module pipe connection types: parallel and serial. Meteorological variable data, the inlet and outlet fluid temperatures, surface temperature, and electrical energy generation of the modules were measured and collected using the measurement equipment according to the module pipe connection type. Consequently, the parallel-type method was approximately 10% more efficient than the serial type in energy production, whereas the serial-type method produced water with a temperature approximately 47% higher than that of the parallel type. Notably, it was advantageous to apply the parallel-type connection to maximize the energy generation efficiency in buildings where the system efficiency is vital and the serial-type connection in buildings where the high temperature of hot water is required.

Visiting patients and medical staff share the same air in outpatient departments of general hospitals. In South Korea, 11 of 15 general hospitals operate with a fixed ratio of 20% outdoor air and 80% recirculated air even during peak visiting hours, causing the recirculation of indoor airborne pathogens. In this study, we propose a scheduling strategy in which air conditioning is adjusted proactively according to the number of visiting patients. The number of visiting patients was measured over five weeks with a people counter. The count showed a regular pattern: an increase in the morning, a decrease at lunch, and a sharp rise in the afternoon. Based on this pattern, we designed a pre-delay schedule for the air conditioning system in which the air intake rate is increased before the increase in visiting patients and maintained for some time after the decrease. The efficacy of this strategy was verified through mock-up room evaluations and full-plane CFD simulations. In the mock-up room tests, the pre-delay operation effectively suppressed particle accumulation in all three conditions. In the CFD simulations, the pathogen count in the waiting room during peak visiting patient hours decreased from the existing 12 particles/10⁶ to 8.4 (−30.0%) with the scheduled operation and 6.8 (−43.3%) with the proposed pre-delay operation. Thus, the proposed method can be applied in existing facilities without modifying an installed central air conditioning system because primarily the air intake rate is adjusted within the airflow range that the existing central air conditioning system can deliver, without adding or modifying equipment.