New Framework for Real-time Measurement, Monitoring, and Benchmarking of Construction Equipment Emissions

• Main Author: Heidari Haratmeh, Bardia
• Other Authors: Civil and Environmental Engineering
• Format: Others
• Published: Virginia Tech 2015
• Subjects: Sustainable Construction; Heavy-duty Equipment Emissions; Portable Emission Measurement System; Vision-based Technology; Discrete Event Simulation
• Online Access: http://hdl.handle.net/10919/64345

The construction industry is one of the largest emitters of greenhouse gases and health-related pollutants. Monitoring and benchmarking emissions will provide practitioners with information to assess environmental impacts and improve the sustainability of construction. This research focuses on real-time measurement of emissions from non-road construction equipment and development of a monitoring-benchmarking tool for comparison of expected vs. actual emissions. First, exhaust emissions were measured using a Portable Emission Measurement System (PEMS) during the operation of 18 pieces of construction equipment at actual job sites. Second-by-second emission rates and emission factors for carbon dioxide, carbon monoxide, nitrogen oxides, and hydrocarbons were calculated for all equipment. Results were compared to those of other commonly used emission estimation models. Significant differences in emission factors associated with different activities were not observed, except for idling and hauling. Moreover, emission rates were up to 200 times lower than the values estimated using EPA and California Air Resources Board (CARB) guidelines. Second, the resulting database of emissions was used in an automated, real-time environmental assessment system. Based on videos of actual construction activities, this system enabled real-time action recognition of construction operations. From the resulting time-series of activities, emissions were estimated for each piece of equipment and differed by only 2% from those estimated by manual action recognition. Third, the actual emissions were compared to estimated ones using discrete event simulation, a computational model of construction activities. Actual emissions were 28% to 144% of those estimated by manual action recognition. Results of this research will aid practitioners in implementing strategies to measure, monitor, benchmark, and possibly reduce air pollutant emissions stemming from construction. === Master of Science