The latest technologies have the potential to reduce CO2 emissions and optimize equipment operation management. Thus, significant advancements have been seen applying advanced tools and technologies. Such developments have opened new venues for construction practitioners and scholars to explore possible outcomes of advanced applications in the construction industry, particularly in the context of sustainability and emissions [26, 27] as there exists enormous opportunities for reducing CO2 emissions in the construction industry by transitioning from traditional methodologies to using more advanced technologies, materials, and adopting more automated procedures in both hardware and software aspects. Hence, a global equipment manufacturing market is currently going through a transformation towards adoption of more innovative technologies such as Internet of Things (IoT), Artificial Intelligence (AI), real time monitoring, machine learning (ML) to optimize construction processes [7]. This endure a gradual improvement within the equipment system such as boosted engine efficiency, lowering emissions, streamlined electro-hydraulic control, and numerous additional breakthroughs [28]. Artificial neural networks (ANN) have the potential to examine intricate correlations that govern the environmental implications of construction activities. In addition, they possess the ability to produce simple models that could be useful in project planning phases. These models optimize process of determining machinery configurations and work plans that can effectively reduce emissions and energy consumption [11].
However, in construction equipment market, efforts on emissions reduction via innovative tools are very limited. Most of the research is centered on exploring CO2 emissions from a small number of on-site vehicles, such as wheel loaders, bulldozers, and excavators. Therefore, efforts must be made to fully adopt cutting-edge solutions. The future studies should add more off-road equipment such as trucks and track loader using real world data collected from projects [29]. The prelaminar synthesis mostly used synthetic data, resulting in a divergence from real-world outcomes. Such results have limitations like constant utilization of machinery while monitoring efficiency during exaction operation. Another drawback is that the proposed models' input data is based on fundamental data from the excavator manufacturer's manual, which disregards unknown scenarios such a complicated idle period when the excavator should be operated [17]. In addition, the calculation of energy and CO2 emission output data relies on indirect measuring techniques. Therefore, it is highly recommended that the calculation of emissions inventories be revised keeping real-world settings [14, 16]. Future investigation must take additional emission test. More operational-level algorithms and models should be designed for other onsite construction machinery like hydraulic excavators, wheel loaders, bulldozer etc. [21, 30]. As equipment down time also adds emission, therefore, to reduce its impact, operator skills must be reviewed for optimal usage of equipment which can lower produced emissions.
Data covering extensive vehicle employment and duty cycle must be recorded to improve future acquisition. This includes data on vehicle lateral movement and usage of equipment like blades and buckets. External parameters like machine operating environment could also impact sustainability. Thus, soil testing can be employed for developing industrial activity-based emission factors. Field studies should describe emission parameters and inventories, evaluate vehicle technology, and assess operating strategies for key action [21]. The real-world duty cycle of on-site machinery at construction site, soil type, digging, and load [12], can offer useful information about operating pattern of equipment to reduce fuel usage and emissions. Further investigation ought to be undertaken to explore the application of different performance efficiencies for excavator fleets to cover a comprehensive range of operating situations often seen in excavation job sites. Engine torque, that significantly impacts actions regardless of earth density and bucket payload conditions, should also be studied to compare the prediction efficiency of the proposed model with engine load factor or engine torque.